Select Page

1. How does a video motion detector work?
2. Describe the principle of remote distributed multiplexing.
3. What are the uses of slip-rings?
4. What is the basic function of video switcher?
5. List the features of video switchers.
6. Differentiate between fixed and speed domes.
7. Explain the feature of picture in a picture available in a quad multiplexer.
8. Name different types of camera housings.
9. Compare the features of indoor housings with those of outdoor housings.
10. What is the role of NEMA in video camera housings?
11. Name different types of indoor housings.
12. Describe the concerns with video motion detectors.
13. Describe the materials and finishing of camera housings.
14. Differentiate between synchronous and non-synchronous video signals.
15. Compare between analog and digital switchers.
16. List the key features of speed-dome systems.
17. Name different types of multiplexers along with their applications.
18. Compare between analog and digital video motion detectors.
19. What is(are) the purpose(s) of dome cameras?
20. Explain in brief how a multiplexer can offer basic video motion detection.
21. List the special features that can be incorporated in dome cameras.
22. Describe the principle of video multiplexer.
23. Name different types of video switcher.
24. What are the typical reasons for false alarms?
25. Define the term motion detection sensitivity.

LOOK AT THE Chapters File
12 pages
Minimum 5-6 sentences each question

Chapter 11

Video Switchers

CONTENTS

11.1

Overview

11.2

Background and Evolution

11.2.1

Small System Switchers

11.2.2

Midsize Systems

11.2.3

Analog Matrix Switcher

11.2.4

Digital Matrix Switcher

11.2.5

Virtual Matrix Switcher

11.3

Small Analog System Switcher Types

11.3.1

Manual

11.3.2

Homing Sequential

11.3.3

Bridging Sequential

11.3.4

Looping-Sequential

11.3.5

Alarming

11.3.6

Synchronous, Non-Synchronous Video

Signal

11.3.7

Switcher Choice

11.4

Matrix Switcher

11.4.1

Analog Technology

11.4.2

Matrix Switcher Control Functions

and Features

11.4.3

Multiple Locations

11.4.4

Digital Switcher

11.5

Virtual Matrix Switcher (VMS)

11.5.1

Evolution

11.5.2

Technology

11.5.3

Remote/Multiple Site Monitoring

11.5.4

Features and Advantages

11.6

Summary

11.1 OVERVIEW

The function of the video switcher, matrix video switcher, and virtual matrix switcher (VMS) in any multiple-camera security system is to connect any camera to any monitor and display the video image in a logical sequence. The

switched camera pictures on the monitor can be recorded on a VCR or DVR, printed on a video printer, or trans-mitted to a remote site. In both small and large installa-tions, the switcher component performs a vital function that simplifies system use and maximizes the information presented to the security operator. In small security sys-tems that have several cameras and one or two monitors, a switcher may not be necessary since all camera scenes can be displayed on several monitors simultaneously. For a medium or large installation (16, 32 cameras, or more), the number of monitors in the control console cannot equal the number of cameras, and a one-to-one, camera-to-monitor correspondence is not practical. Physical space may be limited, and one security guard may not be able to view multiple monitors simultaneously. To view multiple cameras simultaneously on a single monitor, a combiner or splitter, quad or multiplexer is used (Chapters 12, 16).

Analog matrix video switchers cope with the ever-increasing size and complexity of video systems and are used in midsize and large enterprise systems. The essen-tial function of the matrix switcher system is to switch any combination of cameras to any combination of monitors, video recorders, video printers, or transmission channels.

Matrix switchers are based on micro-processor technolo-gies that allow tremendous flexibility in routing and pro-cessing the video signals. These switchers come in various forms including compact, self-contained units that control 16 or 32 cameras, and multiple monitors and keyboards. These compact dedicated switchers include such features as text generation and camera identification. Larger enter-prise systems having hundreds or thousands of cameras and hundreds of monitors are usually based on a modu-lar construction and rack-mounted equipment. Small and midsize microprocessor controlled switching systems can form the central control center for an integrated secu-rity and building management system, combining alarm, access control, fire, and command-and-control functions.

321

322 CCTV Surveillance

An alternative to the self-contained matrix switcher takes the form of hardware added to a PC system. Some medium-size systems (16–64 cameras) use PC boards installed in a standard PC to effect the video switching. The cables for the cameras, monitors, and any other equip-ment (multiplexers, quads, etc.) are connected directly to the PC via RS-232 control cables, simplifying installation and reducing system cost. A disadvantage of this config-uration is the requirement to purchase and maintain a PC as compared with the dedicated microprocessor-based matrix switcher.

In large systems with hundreds of cameras and moni-tors or those requiring multiple control consoles at dif-ferent sites, the approach is to use a PC to control the matrix switcher and other control command functions. The PC is controlled via the microprocessor keyboards connected to the PC using RS-232 or RS-485 protocol. These large systems can connect the images from video cameras to dozens of video monitors, recorders, or print-ers automatically via the RS-232 communication links. They are software-programmable and can simultaneously switch multiple cameras to multiple output devices using salvo switching techniques. Systems like these are very pow-erful and have more functions than can be described here.

For these very large systems, the security operator is confronted with the difficulty of remembering the camera number or the site at which it is located, and how to con-trol it. To overcome this problem, a site plan monitor is provided having maps of the site programmed into it and overlaid with symbols or icons of the cameras and mon-itoring locations. With these visual display units (VDU), the operator can select the camera and area of interest on the map. This is accomplished with input from a mouse or with the operator’s finger if the monitor has a touch screen. This is the ultimate in system control for analog matrix switcher systems. No knowledge of camera or mon-itor number is necessary, and operating the system is as simple as touching the site plan touch screen.

In recent years there has been an evolution of the IP-based VMS that can eliminate some of the shortcomings of the large expensive analog matrix switcher. The VMS can digitally multiplex, switch, record, and transmit the camera signals to the control console monitoring equip-ment and remote locations via LAN, WAN, and wireless LANs (WiFi).

11.2 BACKGROUND AND EVOLUTION

Up until the last few years legacy CCTV surveillance sys-tems have used traditional small switchers, multiplexers, and analog matrix switchers for interconnecting cameras and routing the video signal to monitors, VCRs, video printers, and in large systems to some remote sites via dedicated hardware and cable. These traditional CCTV solutions rely on analog technology and wired cabling to

transfer the video images from the analog cameras to the switchers and onto the monitors and video recorders in the console room. These analog systems are character-ized by long coaxial video signal and control cable runs, simple analog switchers, or large analog matrix switchers to display and record the camera images. The systems have been acceptable in applications where monitoring and recording was only required at a central location and monitoring console. They prove to be expensive when the requirements are for long distances or when the cameras and console room cross public property or inaccessible locations within a facility.

Large systems with many cameras and monitors have incorporated banks of switchers, multiplexers, and large analog matrix switchers to route the camera signals to the appropriate monitoring equipment. The DVR technol-ogy has brought a significant improvement over the ana-log VCR for recording the video camera images and the ability to distribute them to remote locations, to archive them, and to provide rapid retrieval of video image frames recorded at a particular time.

Within the last few years, traditional video CCTV surveil-lance technology is converging with PC and networking technology. This convergence has resulted in the evolu-tion of the VMS, using the digital signal from an IP camera and routing the digital signal to the console display or any other remote site via LAN, WAN, and wireless WiFi digital networks. The result is a dramatic improvement in the features and functionality that can be delivered to the security operator and management at an unprecedented price to performance ratio. The digital video cameras, digital video transmission, and computer networking tech-nology is now revolutionizing the analog video security industry. This new digital technology is entirely computer-based and often uses existing IP infrastructure instead of requiring a dedicated video cabling.

11.2.1 Small System Switchers

One-on-one Display. For a small video surveillance sys-tem with perhaps eight cameras, there can be a one-to-one correspondence between camera and monitor. This means that each camera can be displayed on a single individual monitor. In small systems and when the camera-to-monitor distances are short (a few hundred feet), the switcher and the switching controls are one and the same and are located at the console. In installations having larger dis-tances between the cameras and monitor, the switcher has two separate units with the switcher located near the camera sites and the switching controls located near the console monitor.

Increasing the number of cameras makes it difficult for the operator to effectively view all the monitors and take appropriate action when necessary. Increasing the num-ber of cameras requires that the images from more than one camera be displayed on one monitor. Displaying four

monitors in a quad configuration or 9, 16, or 32 moni-tors on a single display reduces the number of monitors required. The sacrifices are that there is a decrease in resolution and the additional requirement that the oper-ator views many camera scenes on a single monitor. See Section 11.3 for small video system switcher types.

11.2.2 Midsize Systems

One-on-one Display vs. Split Screen vs. Sequencing. A midsize system having multiple cameras and monitors offers the designer a choice of displaying all the cameras on the monitor in a one-on-one presentation, or present-ing multiple camera images on each monitor. When using one monitor it is impossible for security personnel to observe all camera locations simultaneously. If a camera switcher is sequencing from camera to camera there may be a long time delay before a particular camera is seen again. This can leave a gap in the security function.

11.2.3 Analog Matrix Switcher

Analog matrix switchers route multiple analog video sources to multiple video destinations. They can also route audio signals, controls, and other functions from cameras to monitors and analog and digital recorders. The matrix switcher can route composite video, S-VHS, HDTV, RGB, and other video formats. However, a signal type that is input can only be routed to an output of the same sig-nal type. As an example, a composite video input can only go to a composite video output. The analog matrix switch is the workhorse of the industry and the most common interconnect device to connect cameras to monitors, etc.

11.2.4 Digital Matrix Switcher

Most cameras, monitors, recorders, and other functional components of analog video systems are now becoming digital in design. The video system designer has been await-ing the arrival of a digital solution for the analog matrix switch. The digital matrix switch would be a digitized video stream routed to monitors and recorders or other desti-nations in digital form. The truth of the matter is that this scenario of a fully digital matrix switch has not proven effective because the digital matrix switch just doesn’t do enough. It has also not evolved because of the rapid evo-lution and acceptance of high-speed digital transmission over LAN and WAN transmission channels, and the rapid use of the Internet.

11.2.5 Virtual Matrix Switcher

Many of the video surveillance components of analog video systems have become digital and the security system

Video Switchers

323

designer, integrator, and end user have been awaiting the arrival of a digital solution for the analog matrix switch. The digital matrix switch would digitize a video camera stream routed to the monitor, recorder, or other destina-tion in digital form. The rapid evolution of high-speed dig-ital transmission over various transmission channels and the rapid use of the Internet have effectively bypassed the necessity for the digital video switch.

The VMS technology has effectively skipped the digital matrix switch and moved directly from the analog matrix switcher to a VMS that is integrated into the overall security system. The VMS provides full analog matrix functionality using a standard matrix keyboard, but takes advantage of the digital video streams and connections available on LAN, WAN, wireless networks (WiFi), and the Internet. Section 11.5 describes the VMS in more detail.

11.3 SMALL ANALOG SYSTEM SWITCHER TYPES

Small to medium video security systems use five basic switcher types: manual, homing, bridging, looping, and alarming. By using one or a combination of these switcher types, cameras at multiple remote sites can be routed to the security console or multiple monitoring locations for direct observation, recording, or printing. Most sequential switchers, whether homing, bridging, looping, or alarm-ing, have a three-position switch for each camera input. When one of these switches is in the up position, it is said to be in the Bypass mode. Any of the camera switches set in this position will cause the switcher to automatically skip the corresponding camera in the sequential switch-ing cycle. The center position of these switches is called Automatic (Auto) mode. Any camera switch in this posi-tion will cause the switcher to automatically include the corresponding camera in the normal switching cycle. The down position of these camera switches can have several different functions. Where applicable all camera inputs are automatically electronically terminated in 75 ohms by the switchers. The following sections describe the unique features of each switcher.

11.3.1 Manual

The simplest video switcher is the manual switcher, where the console operator manually chooses one camera from a number of cameras and displays the video image on a single video monitor with front panel pushbutton switches, activated manually by the operator to connect the individ-ual camera to the monitor. The manual passive switcher uses a simple switch for contact closure, whereas the manual active switcher uses an electronic switch. Manual switchers are available to switch from 4 to 32 video cam-eras. Figure 11-1 shows the two types available: manual passive and manual active.

324 CCTV Surveillance

MANUAL

CAMERA 1

2

3

4

PASSIVE

o

o

o

MONITOR

SWITCHER

INTERLOCKED

MECHANICAL

75

75

75

75

75 ohm

SWITCHES

TERMINATION

PUSH–BUTTON SWITCHES

MANUAL

CAMERA 1

2

3

4

ACTIVE

o

o

o

SWITCHER

75

75

75

75

MONITOR

AMPLIFIER

INTERLOCKED

ELECTRONIC

ELECTRONIC

ELECTRONIC

ELECTRONIC

ELECTRONIC

SWITCHES

SWITCH

SWITCH

SWITCH

SWITCH

FIGURE 11-1 Manual passive and manual active switchers

11.3.2 Homing Sequential

The homing sequential switcher allows the continu-ous viewing of any normally sequenced video camera (Figure 11-2). The camera signal is connected to a single monitor. This switcher has a three-position switch for each camera: Automatic, Homing, and Bypass. In the Automatic position, the switcher automatically selects and switches the video signal from one camera after another to the monitor according to the sequence set by the security operator. The length of time each camera picture is pre-sented on the monitor (dwell time) can be changed by the operator. The homing sequential switcher automati-cally sequences from one camera to the next, assuming the cameras have not been bypassed. When the specific cam-era control switch is pressed to the Home position, that camera is continuously displayed on the single monitor and the switching sequence stops.

Functionally the three-position front-panel switches on the homing sequential switcher provide three separate camera display functions: automatic sequencing, bypass, and homing (select). When a switch is set to Bypass, that particular camera is not displayed. When the switch is set to Homing, that camera picture is presented continuously on the monitor and in essence overrides the automatic

sequencing function. This permits continuous observation of any particular camera at the operator’s command. In the Automatic position, all cameras are sequenced onto the monitor, one at a time.

11.3.3 Bridging Sequential

The bridging sequential switcher operates like the homing sequential switcher but has the additional feature that two monitors can display the video cameras. Figure 11-3 shows the block diagram for a bridging sequential switcher. Moni-tor 1 always displays the cameras selected for sequential view-ing. Monitor 2 displays only the camera manually selected for detailed viewing. For instance, pressing the switch for camera 1 to the down position puts the picture on the second or bridged monitor for detailed viewing, while the sequence of all cameras not bypassed continues on the first monitor. Monitor 1 sees the switched sequence of cameras while monitor 2 sees a selected camera continuously.

The first monitor (the sequential monitor) functions as a homing sequential switcher. The bridging monitor displays whatever camera is manually selected. This allows the operator to maintain a system overview while viewing in detail the camera covering a scene of particular interest.

Video Switchers

325

CAMERA 1

2

3

4

o

o

o

ELECTRONIC

75

75

75

75

SWITCH

+12 V

AMPLIFIER

ROTATING

SWITCH

AUTOMATIC

A

A

A

BY

B

B

B

1

SELECT

S

S

S

PASS

P

P

P

5

10

AUTO

AUTO

AUTO

AUTO

DWELL

LEVER

TIME

SWITCH

SELECT = HOMING

MONITOR

FIGURE 11-2 Homing sequential switcher

11.3.4 Looping-Sequential

Homing Sequential. The looping-homing sequential switcher operates like the homing sequential switcher, with the additional feature that all camera inputs can be brought out to a second switcher or other device at another location (Figure 11-4). The switcher has the ability to drive a second switcher, monitors, recorders, and trans-mission devices for remote transmission, thereby providing video images at multiple locations for display or recording. Unlike other switchers, the looping-homing sequential switcher camera inputs are not terminated, thereby allow-ing multiple devices to be connected to the switcher output. For proper operation, one of these devices, gen-erally the last device in the line, is terminated in a 75-ohm impedance.

Bridging Sequential. The looping-bridging sequential switcher operates in the same way as the bridging sequential switcher except that the looping feature is added. As with the looping homing sequential switcher, the camera inputs are not terminated in the switcher. Figure 11-5 shows the block diagram for looping-bridging sequential systems. A looping switcher provides the

ability to establish two independently controlled loca-tions. Each station may select any camera for view-ing without interfering with the operation of the other station.

Remote Sequential. The use of the manual, homing, bridging, looping, and alarm versions of sequential switch-ers just described assumes that the distance between the camera location and the monitor (control console) loca-tion is relatively short. In many installations this is not the case and the cost becomes prohibitive to provide separate video coaxial cables from each camera to the distant monitor location. Remote sequential switchers overcome this problem. The remote sequential switcher consists of two parts: a control unit and a switching unit. They are available in all of the aforementioned versions to provide complete system design flexibility. Both units are con-nected by means of multi-conductor cables, fiber-optics, a multiplexed frequency shift key (FSK), or RS-232 commu-nications system (Figure 11-6).

The control unit is located near the monitor, and the switcher unit is located closest to the central location of all the cameras. The physical separation of the switching and control functions avoids the use of individual camera coax-ial cables to the control console. Each switcher requires

326 CCTV Surveillance

CAMERA 1

2

3

4

o

o

75

75

75

75

MONITOR

+12 V

AMPLIFIER

1

ROTATING

MONITOR

SWITCH

AUTOMATIC

A

A

A

AMPLIFIER

2

BY

B

B

B

1

SELECT

S

S

S

PASS

P

P

P

5

10

AUTO

AUTO

AUTO

AUTO

DWELL

TIME

SELECT = HOMING

FIGURE 11-3 Bridging sequential switcher

CAMERA 1

2

3

LOOPING
4 HOMING

SWITCHER

MONITOR

LOCATION 1

1

MONITOR

LOCATION 2

2

FIGURE 11-4 Looping homing sequential switcher

Video Switchers

327

CAMERA 1

2

LOCATION 1

3

MONITOR

MONITOR

4

LOOPING

1

2

BRIDGING

UNTERMINATED

75 ohm

75 ohm

SWITCHER

TERMINATION

TERMINATION

MONITOR

LOCATION

2

2

75 ohm

75 ohm

TERMINATION

TERMINATION

FIGURE 11-5 Looping bridging sequential switcher

only one or two coaxial cables for monitor input. The remote bridging sequential switcher requires two output coaxial cables.

11.3.5 Alarming

An alarming switcher automatically displays a camera image on to a monitor and/or starts a recorder each time it is activated by a camera VMD or other alarm input (Figure 11-7). These switchers are available in homing, bridging, looping-homing, and looping-bridging configu-rations. When an alarm input signal is received, a corre-sponding output signal is generated and transmitted to a monitor, recorder, or printer.

The homing, bridging, and remote sequential switch-ers can be provided with an alarm feature. In the event of an external alarm caused by motion in the video pic-ture and detected by a VMD, an alarm switch closure caused by any type of sensor input, simple switch clo-sure, IR source, or pressure transducer, the alarmed cam-era will automatically override the pre-selected video on the monitor or be automatically displayed on the sec-

ond monitor. When a bridging type switcher is used, the automatic homing of the alarmed camera overrides any manually bridged display on the second monitor. The sequence of all cameras not bypassed continues on the first monitor.

Simultaneously with this switching, an alarm contact within the switcher closes to operate a recorder, video printer, or any other alarm-indicating equipment. Auto-matic alarm-programmed switchers are especially suit-able for applications where monitors are occasionally unmanned and recorders used to record abnormal events. They are also particularly useful during off hours or over weekends when real-time or TL recorders are used to mon-itor multiple cameras.

The output monitoring device can be a bell, light, or other signaling unit, which would notify a security guard to dispatch a guard to the scene or alert a guard at the scene. Even if there are multiple monitors affording the opportunity to observe all locations, the use of alarming switchers puts attention in areas where guard action is really required. The activation of the alarm signals a sig-nificant occurrence within the field of view covered by a particular camera.

328 CCTV Surveillance

LOCATION 1

LOCATION 2

CAMERA 1

MAIN

SECURITY

2

MONITOR

3

1

75 ohm

4

TERMINATION

BRIDGING

SEQUENTIAL

MONITOR

SWITCHER

REMOTE

UNIT

2

SWITCHER

75 ohm

CONTROL

TERMINATION

UNIT

CONTROL CABLE

1

2

3

4 5

6

7

8

TWISTED PAIRS

TWO WIRE MULTIPLEXED

FIGURE 11-6 Remote homing sequential switcher

CAMERA 1

MONITOR

SW1

2

1

SW2

75 ohm

3

TERMINATION

SW3

MONITOR

4

SW4

2

UNTERMINATED

EXTERNAL

ALARM

SWITCHES

ALARM

VIDEO

SWITCHED OUTPUT TO
ACTIVATE VCR OR OTHER

DEVICE WHEN ALARM OCCURS

DVR OR VCR

FIGURE 11-7 Alarming bridging sequential switcher

11.3.6 Synchronous, Non-Synchronous Video Signal

There are two types of video signals that are switched: synchronous and non-synchronous. Synchronous signals lend themselves to methods of switching where controlled transition maintains a degree of signal continuity, and pro-vides a clean, noise-free video picture during switching. Non-synchronous signals involve the inherent discontinu-ity of timing pulses that cannot be corrected by special switching methods, and show up as noise disturbances in the picture. Picture noise in the video signal takes the form of streaks, a momentary black screen, or other picture irregularities. When switching composite video signals, a break may occur during the synchronizing time and the synchronization signal may be completely lost. This results in picture rolling or tearing when the picture from the next camera is displayed on the monitor. The solution to ensure clean video camera switching is vertical-interval switching. With this method, the switching is allowed to occur only during the vertical interval in the video sig-nal between picture frames (Figure 11-8) while no picture

Video Switchers

329

information is being transmitted. Since no visible monitor picture is displayed during the vertical-interval switching time, switching during this period does not cause pic-ture interruption or deterioration. This technique permits switching from one camera to the next with no noise or interruption of intelligence.

To understand vertical-interval switching, refer to Figure 11-9. The camera video signal is generated in the camera sensor. The horizontal camera clocking signal scans from left to right and reads out the video image signal representing the light image on the sensor. When the clocking signal reaches the right side of the sensor, it returns to the left side and begins another scan. During the return time in the analog system, the clocking signal is addressed down to rows of sensor pixels. After it completes 2621/2 scans (one-half of the full frame), the clocking signal reaches the bottom of the sensor and returns to the top. The clocking signal then scans the alternate pixel rows and after completing the second scan the full sensor has been read out. The return time from the end of the last hori-zontal scan to the beginning of the first horizontal scan is referred to as the vertical blanking interval, since during this

PREVIOUS

SWITCHING PULSE

APPROXIMATELY

VIDEO

50 NANOSECOND

SIGNAL

NEXT VIDEO SIGNAL

VIDEO

DURATION

(LAST

(FIRST HORIZONTAL SCAN)

SIGNAL

HORIZONTAL

SCAN)

(X)

FRONT

PORCH

(Y)

BLACK

LEVEL

HORIZONTAL

BACK PORCH

VERTICAL BLANKING (A–B)

SYNC PULSE

(1.1 MILLISECONDS)

VERTICAL

(Y)

SYNC PULSE

(B)

VERTICAL

BLANKING

INTERVAL

PERIOD

EXPANDED

(A)

HORIZONTAL RETURN LINE

HORIZONTAL

(INACTIVE VIDEO)

ACTIVE SCAN LINE

FIGURE 11-8 Vertical interval switching

330 CCTV Surveillance

(A) TWO VIDEO SIGNALS OUT OF VERTICAL PHASE

(B) TWO VIDEO SIGNALS IN VERTICAL PHASE

PHASE

DIFFERENCE

(C) SYNCHRONIZING GENERATOR

(D) CAMERAS WITH PHASE ADJUST

* VERTICAL SYNC PULSES

* EACH CAMERA HAS

SYNC

*

*

PHASE ADJUST

*

GENERATOR

VIDEO

*

VIDEO

*

SWITCHER

SWITCHER

*

*

SYNCHRONIZED

MONITOR

MONITOR

DVR/VCR

SYNCHRONIZED

DVR/VCR

DISPLAY

DISPLAY

FIGURE 11-9 Sequential switching synchronization

time no video signal is generated. In summary, the picture information occurs during the left-to-right scanning and the vertical blanking in-between scans.

In the typical video surveillance application the cameras will not be synchronized. While they may have waveforms or signals like Figure 11-9a, the time relationship between cameras is not synchronized or in phase.

Since the synchronization pulses from each camera occur at different times, when the switcher switches from one camera signal to the next, a noticeably scrambled or distorted non-synchronized image occurs as the monitor tries to adjust to the synchronization pulses of the new signal. A temporarily distorted picture might be tolerable in some simple direct-viewing applications, but in situa-tions where there are multiple cameras or the information is recorded, the result is unsatisfactory. Since VCR and DVR use the camera synchronizing pulses to synchronize the machines, it takes many frames of video for them to synchronize to the new camera signal. During this inter-val, noise or other artifacts are generated each time the

switcher is switched. The out-of-phase signals shown in Figure 11-9a are correctable by at least two methods.

One technique for producing in-phase signals is to install a synchronizing generator that provides a synchro-nizing signal to the cameras and ensures that they are all in the same phase, operating at the same frequency, and syn-chronized (Figure 11-9b). As an alternative, some cameras can be adjusted so that the phase is the same for each cam-era. Phasing each camera to be the same does not produce a clean switchover, however. Even though the signals may be in phase, if the switching occurs during the video por-tion of the signal, there are visible transient effects such as spikes and flashes on the monitor or recorder image. This problem is eliminated by designing the switcher to switch during the vertical interval (Figure 11-8), and hence the name vertical-interval switching.

In operation, the switcher circuitry detects the vertical interval in the signal and delays the actual switchover from one camera to the next, to the time when vertical blanking is occurring. By using this method no transient effects

are visible on the monitor or in the recorded image. The vertical-interval switching technique may not be important in simple systems, but is extremely important in medium to large systems, and in any system using a video recorder.

In summary, the quality of switching, or how smoothly (clear, noiseless picture) the monitor picture from a cam-era 1 can be switched to camera 2, and so on, is influ-enced by two related factors: (1) the type of signals to be switched—synchronous or non-synchronous; and (2) the switching action itself—the time within the video signal in which the switchover occurs.

11.3.7 Switcher Choice

The following summary suggests which switcher to use in small system video applications:

· Passive Switcher. The manual switcher is the simplest and can switch 4, 8, 16, or 32 cameras depending on model, and display any one of them on a single mon-itor. It is available in either passive or active type. In a simple application, any one of the input cameras can be displayed on a single monitor, one at a time, through manual switching by the security guard.

· Sequential Switcher. The sequential switcher is used when it is necessary to switch automatically from cam-era to camera so that the guard can observe all camera scenes sequentially. As in the manual active switcher, the electronic circuitry provides fast, clean switching with no transients on the screen, and is available with camera dwell times of 1 to 50 or 60 seconds depending on the adjustment made by the operator.

· Homing Sequential Switcher. The homing sequential switcher has the additional feature of permitting the operator to stop and look at one particular camera pic-ture continuously or sequentially and display all the camera pictures with a dwell time set by the operator. This system permits the operator to continuously scan through all the cameras and simultaneously pick out one camera and view it continuously. In the sequen-tial mode, the dwell time (length of time any particular camera is viewed) is independently adjustable for each camera. This provides the operator with the flexibility to view different camera scenes for different periods of time. The homing sequential switcher provides the oper-ator with three options and adjustments: (1) automatic switching, (2) timing, and (3) bypass control.

· Bridging Sequential Switcher. The bridging sequential switcher has two separate outputs for two monitors. One output is for the programmed sequence of cameras; the second is for the continuous display of a single cam-era. Unlike the homing sequential switcher, the bridg-ing sequential switcher provides this constant viewing of a selected input without giving up the overview of

Video Switchers

331

all the camera scenes provided by the sequential pro-gram. With the bridging sequential switcher, if the oper-ator wants to observe a particular camera continuously, the operator moves the switch to Select, thereby dis-playing that camera picture on the monitor continu-ously while simultaneously the other monitor continues to display the sequentially switched camera sequence, including the camera that is displayed on the second monitor continuously.

11.4 MATRIX SWITCHER

11.4.1 Analog Technology

Microprocessors, microcomputers, and massive memory solid-state RAM and magnetic hard drives have revolution-ized the video security industry. When a security system has many cameras and monitors and one or more security control consoles in multiple locations, it becomes more efficient to use a configurable microprocessor-controlled video switching and control system called the matrix switcher (Figure 11-10).

A matrix switcher is a means for selecting an input source such as video, audio, or control signals and connect-ing them to one or more outputs. A video matrix switcher is an electronic device that accepts and distributes video signals selected from multiple inputs to multiple outputs.

Many manufacturers produce systems that can switch hundreds (or thousands) of cameras onto hundreds of monitors and recorders. These systems are built in mod-ular form with removable PC boards and rack-mounted modules, permitting the user to begin with a basic sys-tem and expand when necessary. The removable modules and plug-in units are divided into several sub-chassis or modules to provide online serviceability and to reduce or eliminate system downtime. A disadvantage of these sys-tems is that expansion is in multiples of 8, 16, and 32, so that if only one or a few new cameras are planned, only the addition of these larger multiple of cameras is possible.

These switchers have:

· Keyboard and joystick desktop console

· Rack-mounted card cage chassis housing the multiple sub-modules for the switching and control functions

· Remote modules located near the cameras for driving the camera, lens, and pan/tilt hardware, as well as for communicating the information to the control unit

· Power supply.

The initial design of any analog matrix video switching system should begin with a detail schematic diagram of the proposed layout showing camera, control locations, and any other accessory equipment. In addition, a site plan dia-gram should show the distances between equipments and cable routes since many equipments are distance-sensitive.

332 CCTV Surveillance

REMOTE CAMERA LOCATIONS

SECURITY CONSOLE ROOM

REMOTE

EIA 19″ RACK

MONITOR

CAMERA

DRIVERS

1

CAMERA 1

PAN

COMMUNICATIONS

RS-232, RS-485

TILT

RS-232, RS-485

1

LENS

VIDEO

VIDEO

1

2

SWITCHING

2

ALARM

1

CONTROL

2

3

MODULES

N

3

N

PAN

N

TILT

DVR/VCR

POWER

VIDEO

LENS

N

SUPPLY

PRINTER

1

JOYSWITCH

PC

2

LAPTOP

CAMERA PAN/TILT

AUXILIARY KEYBOARD (S)

SECURITY

KEYBOARD

SUPERVISOR

AT OTHER ROMOTE

N

LOGGING

AUXILIARY

LOCATIONS

LINE

DEVICES

PRINTER

CAMERA FUNCTIONS CONTROLLED

FEATURES:

• LENS—IRIS, FOCUS, ZOOM, PRESETS

• ALL SWITCHING FUNCTIONS (HOMING, ALARM, ETC.)

• PAN/TILT—MANUAL, PRESET

• SALVO/BANK SWITCHING (MULTIPLE CAMERAS SIMULTANEOUSLY)

• TIME/DATE, CAMERA ID NUMBER

• PRESET PAN, TILT, LENS FOCUS, ZOOM, IRIS

ALPHA/NUMERIC ANNOTATION

COMMUNICATIONS: RS-232, RS-485

• CABLING: UNSHIELDED TWISTED PAIR (UTP)

FIGURE 11-10 Configurable microprocessor controlled video switching system

The analog matrix control unit contains the system soft-ware and microprocessor hardware. In some systems, cus-tomized switching programs are included in the hardware using electrically programmable memories (EPROM). These solid-state memory devices allow storage of switch-ing instructions to be used at a later time when automatic sequencing is desired. Systems have alpha-numeric charac-ter generators for camera name and location information or other pertinent data. Matrix switchers have text anno-tation card providing each video input with time, date, a three-digit camera ID number, and a multiple-line user-programmable alpha-numeric message display.

Medium- to large-size matrix systems use RS-232, RS-422 or RS-485 transmission protocols for controlling cam-era functions and other output devices. For systems hav-ing up to about 200 cameras and 40 monitors, a single microprocessor-controlled keyboard has sufficient process-ing power to operate the system effectively. One or two slave keyboards may also be added if there is a requirement for more than one person to control the system. Gener-ally, these large video control switching functions are kept separate from any other control functions or other parts

of the security systems such as alarm, access control, fire, and safety.

Communication from the console to the remote con-trol camera module is via RS-232 or RS-485 communi-cation protocol. Distances between the control console and remote console can be 1000–5000 feet, with the data signal cable a single twisted-pair, 22-AWG, shielded wire. Most equipment is housed in 5 to 7-inch-high EIA 19-inch rack-mounted modules, thereby removing most of the electronics from the desktop area except for the keyboard. Some systems have the ability to connect several keyboards to the same control system, thereby permitting control of the system from several locations.

All basic microprocessor-controlled systems have the capability for manual, homing, looping, sequential, auto alarming, bridging, and remote switching functions. A unique feature called salvo switching allows the opera-tor to switch a selected bank of cameras into a bank of monitors as a synchronized group with all of the moni-tors switched together in step. The unique salvo switching feature allows the operator to view all scenes in one gen-eral area, such as a single floor in a building, before switching to the next floor. This feature can significantly

increase the monitoring efficiency of the security guard, since it automatically switches a logical array of cameras.
These systems can provide the same control over alarm functions as over the video network functions. The alarms are constantly monitored by the control console. If one or more of the alarms is activated, the system automat-ically switches in the camera nearest the alarm and dis-plays its video scene on the appropriate monitor. The types of alarm sensors accommodated include switches, infrared sensors (PIR), and VMDs. Alarm signals can be monitored via an audible tone alert or visual indicator. Real-time images can be recorded automatically by hav-ing the recorder switch from TL to real-time recording mode. The operator has the ability to bypass or restore cameras and alarms at will. Individual camera dwell times and sequencing times can be set by the operator on all cameras.

In large systems, the camera, monitor, recorder, and other system functions and hardware are programmed into the PC so that the system can be customized to suit almost any specific security application. System pass-words are programmed and lockout tables used to limit access of unauthorized personnel. In addition to the oper-ational switching sequences normally entered from the PC keyboard, complex switching sequences can be pro-grammed off-line using the PC and then downloaded to the microprocessor control system. Examples of such com-plex switching include pan/tilt presets for camera point-ing position, and lens iris, zoom focal length, and focus settings.

These functions are accomplished via receiving modules located at the camera sites and the RS-232 communica-tions. This function is accomplished: (1) by the operator selecting a specific camera and preset number or (2) auto-matically if the system is preprogrammed, so that when an alarm occurs at a location in the scene, the camera auto-matically goes to the preset condition. Simultaneously, a recorder is activated into real-time mode and records the activity at the designated preset camera position.

Figure 11-11 illustrates a complete matrix switching sys-tem used in a large security application having hundreds of cameras and dozens of monitors, VCRs, DVRs, and print-ers.

All cameras, lenses, pan/tilt platforms, monitors, recorders, and printers are controlled, monitored, and switched via the central matrix switcher. The switcher com-municates control functions to the hardware via RS-232 or RS-485 protocol or time-multiplexed signals. Video signals from the cameras are transmitted from the remote loca-tions via individual coaxial, two-wire, fiber-optic, or wireless channels. The matrix switcher has a separate video input connector for each camera and a separate output connec-tor for each monitor, recorder, or printer device. To bring the matrix switcher and camera and monitoring equip-ment online, it must first be “configured” or programmed

Video Switchers

333

FIGURE 11-11 Microcomputer video switching systems

according to manufacturer instructions, the hardware con-nected to it, and the required functioning of the system. This can take hours or days to accomplish and requires a detailed plan with methodical procedures. Figure 11-12 shows a block diagram of a typical video matrix switcher used in a large security installation.

11.4.2 Matrix Switcher Control Functions and Features

Matrix switchers are supplied with many different control functions and features. Some of these user-defined and fixed controls and features are listed below:

· On-Screen Display: Monitors can display alpha-numeric characters that can be dynamically changed to show camera information such as video input number and title.

· Auto or Manual Sequencing: Camera tours can be pro-grammed for any video output. The security operator may define a dwell time for any video input to create a custom tour.

· Alarm Switching: Alarm inputs can be routed from any input or group of inputs to any video output from a graphical user interface (GUI) or PC.

· System Priority: Keyboard users can be assigned differ-ent levels of security for the control of camera sites. These different levels of access can be granted based on need to know.

· Camera Numbers: Camera IP numbers and names may be assigned to cameras in specific areas around in the facility to better identify camera locations.

· Monitor Numbers: Monitor numbers may be assigned to monitors in different console rooms at a facility or facilities to identify monitor locations.

· Salvo Switching: Banks of cameras may be switched to a bank of monitors with one command.

334 CCTV Surveillance

CAMERA 1

MODULE 1

MODULE 1

MONITOR

PAN/

8 CHANNEL

1 CHANNEL

TILT

8

VIDEO INPUT

VIDEO OUTPUT

RECEIVER

DRIVER

9

MODULE 2

MODULE 2

MONITOR

2

8 CHANNEL

1 CHANNEL

CAMERA 8

16

VIDEO INPUT

VIDEO OUTPUT

PAN/

TILT

RECEIVER

MONITOR

DRIVER

57

MODULE 8

MODULE 8

3

8 CHANNEL

1 CHANNEL

64

VIDEO INPUT

VIDEO OUTPUT

VIDEO

CAMERA 64

VIDEO BUS

CASSETTE

PAN/

RECORDER

TILT

CHARACTER DISPLAY

RECEIVER

VIDEO

SWITCHER

DATA

DRIVER

PRINTER

COMMANDS

CENTRAL

KEYBOARD 1

SIGNAL

REMOTE*

MANCHESTER

PROCESSING

DISTRIBUTION

KEYBOARD 2

COMMANDS

CONVERTER

MODULE

UNIT

COMMUNICATIONS VIA:

POWER

DIGITAL

KEYBOARD 8

MULTIPLEXED

*COAXIAL

SUPPLY

SIGNALS

2–4 WIRE

ALARM

UTP

FIBER OPTICS

INTERFACE

MATRIX SWITCHER

WIRELESS

FIGURE 11-12 Video matrix switcher block diagram

· Partitioning: Password-protected user accounts can be set up with specific access to cameras, sequence tables, multiplexer tables, and salvo tables.

· Camera and platform pan, tilt, zoom (PTZ) Control

· Hardware support for RS-232, RS-422, and RS-485.

Some important capabilities and restrictions a video matrix switcher system should have are:

· Operator should have passwords that allow access to the system.

· The system should have the capability to limit the num-ber of system controllers (keyboards, etc.) that a given operator can log onto.

· The system should have the capability to limit the number of cameras that can be selected by any given operator.

· The system should have the ability to limit the cameras that can be selected or controlled from any operator control location.

· The system should have the ability to limit the monitors that can be viewed from any operator control location.

· The system should have the ability to limit the cameras that can be shown on any particular monitor.

When given access to the system, the operator should be able to form the following basic functions:

· Switch video signals to the monitors

· Operate the camera functions such as pan, tilt, zoom, and focus

· Activate preprogrammed group presets (set groups of cameras to previously selected positions)

· Activate previously established camera tour sequences

· Acknowledge and reset alarms

· Activate auxiliary contacts

· Access camera-specific features by camera menu.

Selected operators should have the ability to program automated sequences as described below:

· Group presets: Ability to set up camera preset positions, including camera to monitor selections.

· Tour sequences: Preprogrammed camera display seq-uences in both forward and backward direction. Each step of the sequence consists of the camera num-ber, dwell time, camera position preset, and auxiliary controlled state.

· Group tour sequences: Multiple camera group presets may be linked together with a dwell time.

The matrix switcher can control many other video equipments such as multiplexers, VCRs, DVRs, quads, motion detectors, and video transmission systems using the RS-232 or other control signals. These RS-232 ports connected to the matrix switcher controller generate the commands and appropriate protocols to operate different functions generated from the keys on the keyboard.

11.4.3 Multiple Locations

Video security systems are often required for large build-ings with many floors with separate guard consoles located away from the main building site or in widely sepa-rated sites.

In large systems with 200 cameras, 40 monitors, or requiring more than two control consoles at different sites, the general approach is to use a PC to control the matrix switcher and other control command functions. The PC is controlled via the microprocessor keyboard connected to the PC using RS-232 or RS-485 protocol. Systems like these are very powerful and have more functions than can be described here. For these large systems, the operator is confronted with the difficulty of remembering the num-ber of the camera, the site at which it is located, and how to control it. To overcome this problem, a site plan moni-tor is provided that has maps of the site programmed into it and overlaid with symbols or icons of the cameras and monitoring locations. With these VDUs the operator sim-ply selects the area of interest on the map and then selects the camera to be used. This can be accomplished with input from a mouse or with a finger if the monitor has a touch screen. This is the ultimate in system control for analog matrix switcher systems. No knowledge of camera, monitor, or monitor numbers is necessary, and operating the system is as simple as touching the screen.

11.4.4 Digital Switcher

Most of the functional components in legacy analog video systems are now becoming digitally networked. The video security industry has been awaiting the arrival of a digi-tal solution for the analog matrix switch. A digital matrix switch would digitize a video signal and route the video stream to the monitor, recorder, or other destination in digital form. The truth of the matter is that this scenario of a fully digital matrix switch has not proven effective because it just doesn’t do enough. It has also not evolved because of the rapid evolution of high-speed digital trans-mission over various transmission channels and the rapid use of the Internet. The technology is effectively skipping the digital matrix switch. Switching systems are moving

Video Switchers

335

directly from the analog matrix switch to the VMS that is integrated into the overall security system.

11.5 VIRTUAL MATRIX SWITCHER (VMS)

11.5.1 Evolution

The VMS provides full analog matrix functionality using a standard matrix keyboard, but takes advantage of the digital video streams and connections available on LAN, WAN, WiFi, and the Internet. The VMS lays a foundation to integrate and enable the combination of three essential security technologies: the DVR, the multiplexer, and the matrix switch.

Matrix switching has evolved from: (1) first-generation video system using a matrix switch, multiplexer and switches, (2) second-generation matrix switch with DVRs connected to the intranet or Internet network, (3) local matrix switching connected to an Ethernet, LAN/TCP/IP switching network, (4) to a true network-based system using a VMS and all Web-based cameras connected to an Ethernet, LAN/TCP/IP switching network with remote access from any location. These four switching systems are shown in Figure 11-13.

Until recently, traditional analog video matrix systems have been the dominant method for routing video signals (Figure 11-14). At the heart of these systems is an analog cross-point matrix switcher that allows any camera input to be viewed on any monitor output. The switchers are usu-ally connected to text generators used to annotate time, date, camera ID, and name information on the displayed video signal. These digital matrix switchers are used with a keyboard and GUI and other devices to control and provide full-featured surveillance functionality.

Legacy analog video systems have some disadvantages in that the video signals are susceptible to external interfer-ence from EMI or RFI noise sources. Coax cables carrying video signals can only be run over distances up to 1000 feet without using optical-fiber or unshielded twisted-pair (UTP) wiring.

Digital networks, on the other hand, deliver signifi-cant advantages over analog transmission methods. These include improved signal integrity over long distances and compatibility with off-the-shelf IT hardware. These net-works allow video surveillance, access control, alarm, and other functions to be successfully routed through LAN, WAN and Internet networks. The network routing of video over these channels is functionally equivalent to the role of the analog cross-point matrix switcher in the legacy matrix system, but is instead distributed through-out the network structure in digital form. In this digital domain, the network replaces the centralized hardware switcher and coaxial cables in the matrix system. Only the keyboards, controller, and text overlays are left intact to preserve the user experience of the legacy matrix system

1st GENERATION SYSTEM

ANALOG CAMERAS

2nd GENERATION SYSTEM

REMOTE PC

ANALOG CAMERAS

CLIENT

SOFTWARE

FIXED

MONITOR

MONITOR

MULTIPLEXER

336 CCTV Surveillance

FIXED

MONITOR MONITOR

FIXED

MULTIPLEXER

FIXED

MATRIX SWITCH

P/T/Z
VCR
CONTROL
PANEL

KEYBOARD PRINTER

3rd GENERATION SYSTEM

ANALOG CAMERAS

REMOTE PC

CLIENT

FIXED

SOFTWARE

MONITOR

MONITOR

MULTIPLEXER

FIXED

P/T/Z

MATRIX SWITCH

NETWORK CAMERAS

DVR

FIXED

ROUTER/

PRINTER

SERVER

INTERNET/

FIXED

CONTROL

INTRANET

PANEL

P/T/Z

ETHERNET LAN/TCP/IP

NETWORK BASED SYSTEM

SUPPORTS IP CAMERAS

SUPPORTS ANALOG CAMERAS

WEB-BASED REMOTE VIDEO ACCESS

MATRIX SWITCH

P/T/Z

DVR

CONTROL

PANEL

INTERNET/

PRINTER

INTRANET

KEYBOARD

MATRIX, ETC.

REPLACED BY

PC WORKSTATION AND

APPLICATION SOFTWARE

4th GENERATION SYSTEM

ANALOG CAMERAS

REMOTE PC

CLIENT

FIXED

SOFTWARE

FIXED

P/T/Z

NETWORK CAMERAS

FIXED

ROUTER/

SERVER

INTERNET/

FIXED

INTRANET

P/T/Z

ETHERNET LAN/TCP/IP

NETWORK BASED SYSTEM

SUPPORTS IP CAMERAS

SUPPORTS ANALOG CAMERAS

WEB-BASED REMOTE VIDEO ACCESS

FIGURE 11-13 Evolution of the matrix switcher to the virtual matrix switch

Video Switchers

337

1

2

3

4

5

6

7

8

1

1

MONITOR

CAMERA 1

2

2

TEXT

2

MONITOR

3

3

CAMERA 2

ANNOTATION

4

4

3

ELECTRONICS

MONITOR

CAMERA 3

5

5

P/T/Z CONTROL

GENERATORS

ALARM EVENT

MONITOR

CROSS-MATRIX

CAMERA N

INPUTS

SYSTEM CONTROLLER
EVENT RESPONSE

GENERATOR

KEYBOARD:

CONTROL

P/T/Z

FIGURE 11-14 Traditional analog video cross-matrix switch

while still delivering the powerful functionality of a full-featured matrix switch. For all intents and purposes, the network represents a cross-point matrix and is in fact a VMS—a virtual video cross-point matrix.

11.5.2 Technology

The first step in realizing the virtual video matrix is to digitize the video signal for transmission over the net-work using a video IP encoder for each analog camera. Figure 11-15 shows the virtual video matrix in which the video signal has been digitized for transmission over the network using a video encoder for each camera.

Internet Protocol cameras already have these encoders built into them specifically to communicate over these networks. The best encoders are designed to supply high-efficiency digital MPEG video streams. Connections for the video and PTZ control signals from each cam-era are made to the encoder using standard coaxial serial data wiring. These encoders also have inputs to support alarm sensor contacts and outputs to control relays or other alarm annunciation devices. Two-way audio is also available as an option.

The next step is to connect the encoder to the nearest network via a Cat-5 or Cat-3 cable. Once video signals are present on the network, there are a number of important security applications that are possible.

Several advantages of VMS technology are realized in any midsize or enterprise security system that is already using computer hardware. There is no need to purchase and install the analog matrix switcher. The requirement and expense for installing coaxial cables or other new wiring throughout a facility is eliminated. The VMS sys-tem allows the user to leverage the computer, monitor, and network that already exists at the facility. Additionally, the hardware is generic so that the end user maintains flexibility and cost control over any new critical hardware decisions.

All analog matrix switch systems have costly and cumber-some scaling limitations. As an example, to add one more monitor to a 32 monitor system requires the addition of shelves of matrix switching equipment, since the systems are based on multiples of 8, 16, and 32 cameras and mon-itors. This is not true of an integrated software-based VMS system. Only additional user-licenses and encoders or IP cameras in the exact increment desired from as small as one to any number of cameras or monitors are needed.

338 CCTV Surveillance

REMOTE SITE 1

REMOTE SITE 2

REMOTE SITE 3

VIDEO: IP CAMERA,

VIDEO

VIDEO

ANALOG CAMERA

ACCESS CONTROL

ACCESS CONTROL

AND SERVER

COMMUNICATION

COMMUNICATION

OTHER

COMPRESSION: MJPEG

OTHER

MPEG-4

ROUTER/SERVER

ROUTER/SERVER

ROUTER/SERVER

NETWORK REPRESENTS

CROSSPOINT MATRIX WITH

ALL DIGITIZED TRANSMISSION

WAN

INTRANET

CABLE: COAXIAL

INTERNET

FIBER OPTIC

LAN

WIFI

UTP

CAT 5

VIRTUAL MATRIX SWITCHER COMPUTER/CONTROLLER

EXISTING IT

COMPUTER

TEXT ANNOTATION

NETWORK

KEYBOARD

CAMERA CONTROLLER

P/ T/Z PRESETS

FIGURE 11-15 Virtual video matrix for network transmission and control

11.5.3 Remote/Multiple Site Monitoring

Enterprise-level systems require customized installations of cable and hardware entailing significant costs and wiring needed to bring analog signals back to the control con-sole, not to mention the distance limitations on these cables. VMS technology eliminates these costly and time-consuming demands. If the user needs to move and relo-cate to a new facility, the cost and miles of wasted coaxial cable represents a major consideration. VMS technology provides the flexibility to meet these demands with min-imum cost and time. Upgrading from an analog to VMS does not extend the availability of video information. How-ever, when either the analog or a digital matrix system is integrated into a networked video system, the VMS system provides wide area connectivity, and video data becomes available anywhere. The VMS technology allows organiza-tions to fully leverage their security investment. With the level of access and functionality provided by the VMS sys-tem: (1) human resources now have a visitor monitoring system, (2) operations have the ability to monitor traf-fic in the lobby or loading dock areas or elsewhere, (3) retail operations can prevent overloading at cash register lines and can monitor cashiers, and (4) marketing per-

sonnel can remotely monitor the level of interest shown at product displays in stores. Many other applications can be cited.

11.5.4 Features and Advantages

Analog video systems require a dedicated wiring and cabling for each camera. Digital systems using VMS tech-nology require only Cat-3, Cat-5 cables or digital wireless transmission. The VMS with built-in DVR capability can record video images without any degradation loss, sup-port multiple playback, re-recording, and transmission, and can distribute the video images to multiple locations. The VMS represents a centralized control and record-ing ability allowing local monitoring and remote multiple site viewing. The video camera generates a digital signal using digital signal processing (DSP) in the camera and produces a digital signal at the output. It transmits the dig-ital video signal over the LAN, WAN, or wireless network while retaining complete integrity and image quality. The VMS likewise distributes and records the digital signal so that it remains high quality during switching, reproduc-tion, and transmission.

The digital video system with VMS produces evidence that has high integrity. When producing evidence from a standalone DVR there is no way to verify the actual source of the images as cameras can be switched on the back of the DVR unit. Using IP cameras and the VMS, the images are kept under the MAC address of the specific camera. This is a clear one-to-one identification of the source of images.

Since the VMS uses off-the-shelf servers, workstations, and computers, the system can always be upgraded to the latest hardware for the best price/performance. This is also true of the IP cameras and other software and hardware that support the system. The VMS system can integrate existing analog cameras, infrared (IR) cameras, covert cameras, and of course the IP camera. The digital technology permits object recognition and tracking, face

Video Switchers

339

recognition, license plate recognition, direction detection, people/car counting, etc.

11.6 SUMMARY

The heart of a good security system is a highly functional video switching control system. In small systems, the switch-ers will take the form of simple passive, homing, sequential or alarming switchers. In medium- to large-size systems, the switchers will take the form of an analog cross-point matrix switcher or a VMS. During the design phase of any security system, management and security personnel must decide what information needs to be displayed, acted upon by the security operator, recorded, and printed, and choose the switching system suitable to accomplish the task.

Chapter 12

Quads and Multiplexers

CONTENTS

12.1

Overview

12.2

Background

12.3

Quad Split-Screen Displays

12.3.1

Quad-4 Image

12.3.2

Multi-Image 9, 16, 32

12.4

Multiplexer Technology

12.4.1

Image Rate vs. Number of Cameras

12.4.2

Encoder/Decoder

12.5

Hardware Implementation

12.5.1

Simplex

12.5.2

Duplex/Full Duplex

12.5.3

Triplex

12.6

Recording and Playback

12.6.1

Analog and Digital Recording

12.6.2

Video Playback

12.7

Video Motion Detection

12.8

Alarm Response

12.9

Integrated Multiplexer and DVR

12.10

Remote Distributed Multiplexing

12.11

Summary

12.1 OVERVIEW

Sequential switchers display the images from multiple cameras on one monitor sequentially, one at a time, with a dwell time between the display of each camera image. A disadvantage of sequential switching and recording is that when a single video camera is displayed on the monitor all the other cameras are not being viewed. This can result in a great loss of intelligence from the cameras not being displayed. Each camera image is displayed for a dwell time set by the operator adjusted from a few seconds to many seconds. With the use of sequential switchers, many activ-ities on many cameras can be missed since not all camera scenes are being displayed simultaneously.

A quad or video multiplexer displays all of the images from many cameras onto a single split-screen monitor simultaneously. These devices generate a video signal that can record all the images at a much higher refresh rate than is possible with a sequential switcher. The use of video multiplexers eliminates the normal video time gaps created by conventional sequential switchers.

There are basically three generic types of multiplex-ers: simplex, duplex/full duplex, and triplex. The simplex multiplexer can display multiple images—4, 9, 16, and 32—on the same multi-screen monitor. The duplex mul-tiplexer displays multiple images on a display but can also provide the necessary encoding and decoding signals to simultaneously record images on a VCR or a DVR. A triplex multiplexer can simultaneously display multiple live images on a display, record camera images on a recorder, and display playback images from a recorder.

Most multiplexers offer some form of basic video motion detection (VMD). This might be listed in a variety of ways in the literature but it essentially amounts to detecting movement in the field of view of the camera by electroni-cally discerning changes in the light level within the image.

In addition to displaying motion, multiplexers can respond to alarm inputs from external sensors (door switch, infrared detectors, glass break, microwave, etc.). Manufac-turers are quick to point out, however, that the multiplexer’s primary purpose is to furnish efficient video multiplexing and multi-screen display. Alarm handling and motion detec-tion are secondary functions, and the video multiplexer sys-tem should not be the only alarm device on site.

12.2 BACKGROUND

Video multiplexing is an example of time division multi-plexing. The video multiplexer constructs a sequence of pictures captured from each of a number of cameras, in

341

342 CCTV Surveillance

CAMERA

1

2

3

8 CHANNEL

4

MULTIPLEXER

MONITOR

5

6

7

8

1

8

1

9

16 CHANNEL MULTIPLEXER

DVR/VCR

FIGURE 12-1 Video multiplexing system diagram

turn, and displays the video images in a split-screen format on one monitor (Figure 12-1).

The initial electronic image splitters became available in the form of a four-way splitter or quad. Subsequently the 9, 16, and 32 camera image splitters—multiplexers— became available. These larger units had the ability to take synchronized or unsynchronized cameras and display them on a single monitor simultaneously in a synchronized and stable format.

Early multiplexers were basically video switchers that could mark each camera with a unique ID number in the vertical interval. This required the cameras to be gen-locked or v-phased (vertical sync) so the VCR would see a contin-uously composite sync signal and so that it would not lose servo-lock on the switched incoming video signals. To play-back the camera images, the VCR switched the correct cam-era onto its output only during its active period on the tape and switched to a gray solid background picture for the rest of the time. This caused severe image flicker but produced a viewable single camera image display and was effective. Later generations of this multiplexer design saved the active camera image until a new picture was displayed eliminat-ing the gray background, providing a better playback result. The primary benefit of this technology was that this device

guaranteed a continuous composite sync to the VCR regard-less of the video signal quality. A secondary benefit was that a non-gen-locked or any other camera could be used with this multiplexer. Present-day cameras have quality gen-locking systems, and/or stable line-locked vertical interval sync, and DVRs are used so this is no longer an issue.

The quad or multiplexer can also output these pictures as a single continuous video signal with all the necessary encoding and decoding for recording on a VCR or DVR or network. The multiplexer adds the digital camera ID coding to the signal so that the individual camera fields belonging to each camera can be identified and recovered by the recording equipment on replay. For display moni-toring purposes, the same sequential scan process is used and each camera’s picture used is electronically reduced in size and displayed in a pre-determined position on the screen. Each camera is assigned a different position so as to produce the familiar mosaic or cameo of reduced size camera images on a single display monitor.

Most current multiplexers have the ability to display 4, 9, 16, or 32 pictures simultaneously on the screen. Usually the image update rate is the same as the output multiplex rate, but some manufacturers have refresh rates up to real-time capability. This feature is useful since the screen is for

viewing only the multiplexed output being a full screen high-resolution image. The multi-screen feature to display an alarmed camera in a cameo format is very useful in playback of all recorded images, and later single camera selection when an alarm or some other activity needs to be viewed.

12.3 QUAD SPLIT-SCREEN DISPLAYS

The quad display is the simplest form of this multiplexing technique where the signals from four cameras are pro-cessed to appear in four quadrants of a single monitor display.

12.3.1 Quad-4 Image

The quad splitter permits viewing four live video cam-eras simultaneously or selecting one camera full screen or sequencing through all or selected cameras (Figure 12-2).

The quad splitter can display the images in quad or full screen format while recording to a VCR or DVR in quad format. On playback from the recorder the image from the quad multiplexer can be zoomed up 2X and a freeze frame image can be displayed for detailed analysis. The resolution of the quad ranges from 720 × 480 pixels up

Quads and Multiplexers

343

to 1024 × 512 pixels for high-resolution systems. The units have the ability to annotate the video image with time, date, camera ID, and title in both the live monitor display and recorded image display. The quads provide 30 fields per second, real-time refresh rate.

Figure 12-3 shows diagrammatically the different scene formats that the quad system can display. In the single camera select mode the full screen images from camera 1, camera 2, camera 3, and camera 4 outputs can be selected. In the quad mode, four camera scenes are displayed and each individual picture on the monitor is a full camera scene reduced in size (compressed). Many quad systems can “freeze” a displayed image on the monitor for detailed examination. This permits the security operator to view a single scene in more detail over a period of time until it is released by the operator. In this mode, a recording can be made of the full screen or quad pictures.

Other options include alarm mode, video loss indica-tion, and security lock. In the alarm mode, the system brings the alarmed camera to full screen on the moni-tor alerting the operator of an alarm activity, all while the recorder records in quad format. Another feature available in some quad multiplexers is called picture in a picture (PIP) in which a reduced image from one cam-era is embedded into the full screen image of another camera.

1

2

3

4

SCENE 2

SCENE 3

SCENE 1

SCENE 4

1

2

1

2

3

4

3

4

QUAD

SELECT

SEQUENCE

MENU

1

2

3

4

· 2

· 4

MONITOR DISPLAYS QUAD PICTURES OR ANY INDIVIDUAL SCENES IN THREE MODES:

QUAD—FOUR COMPRESSED PICTURES SELECT— ONE FULL PICTURE SEQUENCE THROUGH 4 SCENES

FIGURE 12-2 Quad splitter display block diagram

344 CCTV Surveillance

QUAD MODE—COMPRESSED SCENES

SEQUENCE MODE—FULL PICTURES

SCENE 1

SCENE 2

SCENE 1

SCENE 2

SCENE 3

SCENE 4

B

SCENE 4

A

SCENE 3

B

A

t = T1

t = T2

t = T3

t = T4

SELECT MODE—FULL PICTURES

SCENE 1

SCENE 2

SCENE 3

SCENE 4

B

OR

A

OR

OR

FIGURE 12-3 Quad combiner system

12.3.2 Multi-Image 9, 16, 32

Multiplexers are available to display images from 9 to 32 cameras on the same monitor display and are available with similar features to those found in the quad system. Figure 12-4 shows two 9 and 16 camera examples of these systems and the monitor images. Table 12-1 lists features of some of the quad split-screen equipment available.

12.4 MULTIPLEXER TECHNOLOGY

A quad or multiplexer is an electronic device that time-multiplexes video pictures from many cameras onto one video display or video recorder. This means that the mul-tiplexer displays one field or one frame from one camera, and then immediately following that picture it displays the field or frame from the next camera. It repeats the same procedure for all subsequent cameras, and then starts all over again. These images of multiple cameras are displayed on one monitor simultaneously. Using this technique the full resolution of each camera is maintained but the dwell time between displayed or recorded image—2–3 second dead time switching time from a sequential switcher—is reduced instead to milliseconds. When recording the camera sig-nal to a VCR or DVR, the multiplexer switches its input circuitry to each of the connected cameras, in turn. To syn-chronize the cameras during recording, a series of digital codes are embedded into the multiplexer output signal. Part of this code identifies the camera channel number so that the channels may be electronically recognized by the multiplexer during playback. During playback,

another part of the code carries alarm status informa-tion so that external alarm events are also recorded on the tape.

Time division multiplexing combines several camera video input signals into one video output signal to dis-play all the camera images on the monitor simultane-ously. Single images are digitally captured from each of the video input channels, and then lined up (queued) sequentially to form a continuous video signal of time-sliced camera images. Included with each captured image of video can be status information such as alarms, cam-era titles, and time/date. Captured images are controlled by an internal library that the multiplexer automatically modifies to respond to alarms, motion detection, or video loss (Figure 12-5).

To generate a multi-picture mosaic display, the multi-plexer switches its input circuitry to each of the connected cameras, in turn. The multiplexer has a video frame store (and electronic image memory) used to capture a single picture from each camera. As each image is captured, its size is electronically reduced by a predetermined factor and the resulting cameo picture is written into part of the frame store. This results in a small image from the selected camera channel being frozen in one area of the display screen. The same process is provided for each of the cam-eras to similarly reduce the size and to position the image in a particular location of the monitor display area. As the multiplexer scans repeatedly around the channels, each image is continuously refreshed and updated with new images from the designated camera. This results in the familiar mosaic of small camera images.

Quads and Multiplexers

345

9 CAMERA

16 CAMERA

COMBINER

COMBINER

CAMERA 1

CAMERA 9

CAMERA 1

CAMERA 16

MODES:

MODES:

SELECT (FULL)

SELECT (FULL)

QUAD

1

9

QUAD

1

16

NINE

NINE

SELECT

SELECT

9

SIXTEEN

1

2

3

4

5

6

7

8

1

2

3

4

5

6

7

8

9 10 111213 1415 16

QUAD

NINE

QUAD

NINE

SIXTEEN

FULL SCENE

1

2

3

FULL SCENE

1

2

3

4

FROM EACH

FROM EACH

5

6

7

8

CAMERA

4

5

6

CAMERA

9

10

11

12

(COMPRESSED)

(COMPRESSED)

7

8

9

13

14

15

16

(1/3 × 1/3 = 1/9)

(1/4 × 1/4 = 1/16)

MONITOR

MONITOR

FIGURE 12-4 Multi-image 9, 16, split-screen display

EQUIPMENT TYPE

FULL

SCREEN

4 CHANNEL—MONOCHROME

STANDARD RESOLUTION

4 CHANNEL—MONOCHROME

HIGH RESOLUTION

4 CHANNEL—COLOR

STANDARD RESOLUTION

4 CHANNEL—COLOR

HIGH RESOLUTION

STANDARD FEATURES ON MOST QUADS:

GRAY SCALE—256

DIGITAL ZOOM—2x

ADJUSTABLE SEQUENCE: 1–120 sec REMOTE CONTROL: RS232, 422, 485 ALARM-DRY CONTACTS, RS232, 422, 485 4-ALARM INPUTS

OPTIONS: TIME/DATE ANNOTATION CAMERA ID ANNOTATION

SCREEN DISPLAY MODE

RESOLUTION–FULL SCREEN

GRAY

QUAD

(H ×V)

COLORS

SEQUENTIAL

LEVELS

(4 CAMERA)

NTSC

CCIR/PAL

512 × 512

648 × 512

64

1024 × 512

720 × 576

256

720 × 480

64

16 M

1024 × 512

256

16 M

SCREEN FREEZE CAPABILITY

VIDEO LOSS ALARM

LOOP THROUGH TO DVR/VCR FOR RECORDING AND PLAYBACK

SETUP MENU, ENGLISH, OTHER

NTSC 525, PAL 625 TV LINES

Table 12-1 Quad Split-Screen Equipment Parameters and Features

346 CCTV Surveillance

CAMERA REPEATS EVERY 4 FIELDS

1

2

3

4

1

2

3

4

1

CAMERA 1

CAMERA 2

CAMERA 3

CAMERA 4

CAMERA 1

CAMERA 2

CAMERA 3

CAMERA 4

1

2

3

4

5

6

7

8

9

FIELDS

FOR THE MULTIPLEXER WITH 4 CAMERAS EACH CAMERA IMAGE REPEATS EVERY 4 FIELDS = 4 × 1/30 sec = 0.133 sec

FIGURE 12-5 Multiplexed signal from video stream

Most multiplexers can display the video cameras in four different configurations: (1) quad, 4-way, (2) 9-way,

· 10-way, and (4) 16-way, and of course full screen for any camera. Many can also display the cameras in different size configurations. Figure 12-6 illustrates some of these split-screen presentations.

In a standard sequential switcher the camera images are displayed at a 30 frame per second rate. They are displayed sequentially on the monitor at a rate determined by the number of cameras in the system and the pre-assigned dwell times for each camera. In the multiplexer switching system the number of images displayed per second is based on the total number of camera inputs.

If there is only one input the multiplexer displays at a

30 fps rate whereas with four camera inputs it display at

a 7.5 images per camera rate. With a larger number of cameras, say 16 camera inputs, the final display rate would only be approximately two images per second per camera, producing a very jerky display (Figure 12-7).

Multiplexers now feature RS-422 and RS-485 and over-the-coax digital PTZ control to eliminate the need to provide additional controlling units for camera platform pointing and lens control. Other features include motion detection, electronic digital zoom, adjustable image sizes, and RS-232 interfaces to other equipments.

The most common type of camera identification is the annotation of digital information into the vertical inter-val time of the video signal. This is accomplished by dividing a line of video into, say, eight different sections. Each section is defined as a one or zero by either the

1 2

3 4

4 IMAGES 2 × 2

1 2

10

10 IMAGES 2×8, 1×2

1 2 3

4 5 6

7 8 9

9 IMAGES 3 × 3

1

16

16 IMAGES 4 × 4

FIGURE 12-6 Multiplexer multi-screen displays

Quads and Multiplexers

347

FIGURE 12-7 Multiplex camera

CAMERA FIELD SEQUENCE
sequencing technique

A1

A2

A3

A4

A5

A6

A7

A8

CAMERA A VIDEO FIELDS

B1

B2

B3

B4

B5

B6

B7

B8

CAMERA B VIDEO FIELDS

C1

C2

C3

C4

C5

C6

C7

C8

CAMERA C VIDEO FIELDS

D1

D2

D3

D4

D5

D6

D7

D8

CAMERA D VIDEO FIELDS

MULTIPLEXED CAMERA FIELD SEQUENCE

A1

B2

C3

D4

A5

B6

C7

D8

MULTIPLEXED VIDEO FIELDS STREAMED TO DVR OR VCR

presence or absence of black video or white video. By doing this with eight sections it can be interpreted as one byte of digital data that can be converted to a number from 0 to 255.

12.4.1 Image Rate vs. Number of Cameras

A factor to be considered is that multiplexers are basically fast video switchers. When many cameras are connected and the time lapse (TL) recording is too slow, the time of recording a single image from a particular camera may be too long to catch any event. The multiplexer system basically takes the number of camera inputs and dividing that by the recorded pictures per second. To calculate the refreshed or update rate:

Update rate =

Number of Cameras

(12-1)

Recorded Pictures/second

If there are 16 cameras in the system and the recording time is 168 hours in TL mode, then it will take 17.4 sec-onds to record a new image from any one camera input. This obviously would have little use in any application since someone could walk by and never be recorded. Reducing the recording time to 24 hours in TL, there would be a new image every 3.2 seconds. This would be more accept-able but not very applicable in high-traffic areas. Going to a 24-hour virtual real-time (pictures per second), there

would be a new image every 0.8 seconds. This would be useful in most applications.

12.4.2 Encoder/Decoder

Multiplexers require encoders and decoders to identify each of the incoming video camera signals for processing. All current encoder/decoder designs use analog to digital (A/D) converters to convert the standard video signal into a digital format for use with common digital logic devices. After the signal is processed, it is later converted from digital to analog (D/A) for output to be displayed back onto the analog video monitor or recorder.

12.5 HARDWARE IMPLEMENTATION

There are basically three different generic types of mul-tiplexers: (1) simplex, (2) duplex/full duplex, and (3) triplex (Figure 12-8).
The simplex multiplexer can display multiple images: 4, 9, 16, and 32—on the same multi-screen monitor. The duplex multiplexer displays multiple images on a display but can also provide the necessary encoding and decod-ing signals to simultaneously record images on a VCR or a DVR. A triplex multiplexer can simultaneously display mul-tiple live images on a display, record camera images on a recorder, and display playback images from a recorder.

348 CCTV Surveillance

CAMERA 1

CAMERA 1

CAMERA 1

DUPLEX AND

SIMPLEX

CAMERA 2

CAMERA 2

TRIPLEX

CAMERA 2

FULL DUPLEX

MULTIPLEXER

MULTIPLEXER

CAMERA 3

MULTIPLEXER

CAMERA 3

CAMERA 3

CAMERA 4

CAMERA 4

CAMERA 4

MONITOR

DVR/VCR

DVR/VCR

MONITOR

MONITOR

DVR/VCR

DVR/VCR

MONITOR

DVR/VCR

DISPLAYS SINGLE OR MULTIPLE IMAGES ON MONITOR

CANNOT RECORD AND SHOW MULTISCREEN DISPLAY AT THE SAME TIME

DISPLAYS THE MULTISCREEN

AND RECORDS ON THE DVR/VCR

SOME CAN PLAYBACK FROM ONE

RECORDER WHILE RECORDING ON

ANOTHER AND GIVE UP THE MULTI-

SCREEN VIEWING IN THIS MODE

MONITOR DVR/VCR

ALL THE FEATURES OF THE FULL DUPLEX. IN ADDITION THE MULTI-SCREEN OUTPUT CAN BE SUBSTITUTED FOR A THIRD DVR/VCR RECORDER

NOTE: 1. THE FULL DUPLEX CAN RECORD THE MULTIPLEXED OUTPUT TO ONE DVR/VCR, PLAYBACK FROM ANOTHER AND VEIW THE MULTISCREEN AT THE SAME TIME.

2. MONITORS DISPLAY EITHER FULL SCREEN OR MULTISCREEN.

FIGURE 12-8 Generic multiplexer types: simplex, duplex, and triplex

12.5.1 Simplex

The simplex multiplexer is the lowest-cost multiplexer type, has the least number of features, and is easy to install and set up. They are generally used in small systems when there is no security operator active at the console. The simplex multiplexer does not have the ability to record and show a multi-screen display and record simultaneously. The simplex multiplexer unit can either display or record the video information with the initial setup of the mul-tiplexer determining the choice. They are available for monochrome or color camera systems and with options for VMD and alarm handling.

12.5.2 Duplex/Full Duplex

A duplex multiplexer is designed to display either:

(1) a live camera view, (2) a live multi-screen display, or (3) previously recorded images. This multiplexer has the ability to display the multi-screen camera images and record the multiplexed video and control data to the VCR or DVR. Some duplex multiplexers can playback from one recorder while recording on another but the multi-screen viewing is forfeited.

A full duplex multiplexer has the ability to: (1) record the multiplexed output to one recorder, (2) playback

from another, and (3) view the multi-screen at the same time. Duplex and full duplex multiplexers are available for monochrome or color camera systems and with options for VMD and alarm handling.

12.5.3 Triplex

A triplex multiplexer allows viewing of live and recorded images on one monitor simultaneously, eliminating the need for a separate playback monitoring station. The triplex multiplexer has all the features of the full duplex but the multi-screen output can be used for a third recorder, or to display live video which is the more com-mon application (Figure 12-8).

Triplex multiplexers are available with two monitor outputs. Output #1 produces a full screen or multi-screen digital image display that can be frozen on the screen or zoomed in or out. Output #2 displays a full screen, live.

Triplex multiplexers are available for monochrome or color camera systems in 10 and 16 camera models with options for VMD and alarm handling. They have on-screen menu prompts to simplify installation and setup. Table 12-2 lists features of some of the multiplexer equip-ment available.

Quads and Multiplexers

349

EQUIPMENT TYPE

MULTISCREEN

RECORD TO

RECORD TO

RESOLUTON

DVR/VCR,

NUMBER OF CAMERAS *

DISPLAY

DVR/VCR

NTSC

PAL

DISPLAY

SIMPLEX**:

720 × 512

720 × 512

16 CHANNELS

32 CHANNELS

DUPLEX:

4, 8, 16

16 CHANNELS

720 × 480

720 × 576

32 CHANNELS

4, 8, 16, 32

FULL DUPLEX:

4, 8, 16

720 × 572

720 × 572

16 CHANNELS

32 CHANNELS

4, 8, 16, 32

TRIPLEX‡:

4, 8, 16

720 × 512

720 × 512

16 CHANNELS

4, 8, 16, 32

32 CHANNELS

· 4 CHANNEL AND 10 CHANNEL ALSO AVAILABLE

· MOST SIMPLEX DO NOT HAVE CAPABILITY TO DISPLAY MULTI-SCREENS

· DUPLEX—SOME CAN PLAYBACK FROM ONE DVR/VCR WHILE RECORDING ON ANOTHER BUT GIVE UP MULTI-SCREEN VIEWING DURING THIS TIME.

· TRIPLEX—ALL FEATURES OF DUPLEX BUT MULTI-SCREEN OUTPUTS CAN BE SUBSTITUTED FOR A THIRD DVR/VCR.

Table 12-2 Multiplexer Equipment Parameters and Features

FEATURES IN MOST:

ALARM INPUT: EACH CAMERA ALARM OUTPUT ANNUNCIATION DIGITAL ZOOM: 2x DATE/TIME, CAMERA ID

VIDEO LOSS INDICATION

ON-SCREEN MENU

12.6 RECORDING AND PLAYBACK

12.6.1 Analog and Digital Recording

When recording the camera signal to a VCR or a DVR, the multiplexer switches its input circuitry to each of the connected cameras, in turn. The video frame-store in this mode is used to capture a single full-screen field image from each camera. The separate video fields captured from the cameras are re-synchronized by the frame-store for recording onto the video recorder. Using this method, it is possible to record at an average rate of more than 30 cameras per second. Since the video frame-store inher-ently time-base corrects the sync and synchronizes the camera signals as part of the camera capture process, the cameras need not be externally synchronized.
To synchronize the camera signals during recording, a series of digital codes are embedded into the multiplexer output signal. Part of this code identifies the camera channel number so that the channels may be electronically recog-nized by the multiplexer during playback. During playback, another part of the code carries alarm status information so that external alarm events are also recorded.

12.6.2 Video Playback

When the multiplex recording is played back through the multiplexer to the monitor, the multiplexer first extracts the digitally coded data element and uses this to identify the camera ID number information. When a valid channel

number is identified, the multiplex captures the associated video images in the frame-store. The simplest playback mode is where a single camera channel is requested for playback. In this mode the multiplexer captures the cor-responding video images from this camera and displays them as full screen images, and updates each time it iden-tifies another image with the same camera ID number. The embedded digital data packets are decoded, and all the associated status information, titles, time and date of recording, etc. are re-constructed and displayed with on-screen text during playback.

During playback, the user can select one of several screen formats, the cameras to be displayed, and the cam-era positions in the multi-screen display on the unit. The playback speed is selected on the recorder, not on the multiplexer unit.

For multi-picture playback mode the frame-store is used in a similar way as that used for live multiplexer viewing. Many channels are reviewed off the recorder alongside one another on the screen. In this mode the same size reduction and image positioning processes go on in the multiplexer as was described for the live multi-picture mode. In this case, however, there is only one input sig-nal, i.e. that coming from the video recorder playback, but there are many camera channels within the signal. The multiplexer recognizes each new camera channel number, it size-reduces the captured image, and places it on the screen in the predetermined location and with a size cor-responding to that camera. This results in a multi-picture display very similar to the live one showing recorded infor-mation rather than live video. If the operator sees an

350 CCTV Surveillance

image in the cameo images of interest, the multiplexer can be switched to the full screen mode for that camera, to examine the scene more closely.

12.7 VIDEO MOTION DETECTION

Most multiplexers offer some form of basic motion sens-ing or VMD. This might be listed in a variety of ways in the literature but it essentially amounts to detecting move-ment in the field of view of the camera by electronically discerning changes in the light level within the image.
In operation the multiplexer digital signal processing (DSP) electronics determines if something has changed in the video image of any camera. If nothing has changed the multiplexer records fewer pictures per second from that camera, thereby increasing the images per second recorded for other cameras that have motion or activity and their scenes. One caution regarding VMD: some out-door environments have complex detection requirements. In those cases, use a non-video motion detector sensor intended specifically for such situations.

Motion sensing can be considered effectively as an alarm that is flagged internally to the multiplexer. This feature is particularly useful in the recording mode since it can

allow the frame/field recording rate of the recorder to be altered such that images showing movement are recorded at a faster rate than the static ones.
The multiplexer can optimize the display and recording by displaying and recording only video camera images in which activity is occurring.

12.8 ALARM RESPONSE

Multiplexers offer VMD including built-in zone selection with sensitivity settings and alarm linking per camera. Motion detection is used to adjust the rate at which camera images are recorded and can also act as an intrusion alarm sensor to trigger on alarm input. The VMD can be used to simply optimize recording or as an alarm condition. Motion detection can be used as an alarm condition only if movement is detected where no movement is expected or permitted.

Non-video alarms are signal inputs from external sensors that can be acted upon by alarm monitoring hardware or a security operator at a console (Figure 12-9).
Common alarm sensors take the form of door con-tacts, PIR, glass break, microwave motion sensors, trip-wire, photo-electric, magnetic, seismic, etc. All are examples of

ALARM SENSORS:

DOOR CONTACT

PIR MOTION SENSOR

CAMERA 1

MICROWAVE MOTION

2

PHOTO-ELECTRIC

GLASS BREAK

3

SMOKE/FIRE

4

MULTIPLEXER

MONITOR

8

1

1

9

DVR/VCR

TO ALARM

ANNUNCIATOR

FIGURE 12-9 Alarm signals trigger multiplexer

external devices that can output signals to the multiplexer when an intruder enters a monitored area, and be used for alarm annunciation. All these devices can be used by most multiplexers as an input to bring up the picture of the camera that is located in the alarm sensor area and to annunciate an alarm via sound or light indicator. They are also used to command the VCR or DVR recorder to change from TL to real-time recording mode for that camera, and record at a faster speed. In normal use, the recorder is in TL mode to make economic use of the storage media. When an alarm event occurs, the recorder speed is increased to real-time. State-of-the-art multiplexers can cope with making these changes from TL to real-time and acting on alarm inputs. An input from an external alarm by a contract closure to the recorder or by a serial RS-232 port command will be multiplexed, though, and will cause the recorder to change speed. The multiplexer also makes it possible to select logical groups of cameras and to salvo or bank switching of those cameras. Salvo switching accomplishes the switching of several or many cameras in a related zone simultaneously when an alarm input occurs. As an example, in a 20 camera installation the normal record-ing set up may provide the TL recording for all 20 cameras. There may be PIR, other motion sensors, and/or switch sensors in the area. When an alarm event is triggered via one of these sensors, the multiplexer causes the images from cameras in the area of the sensors to be recorded in real-time. Ideally the video system should take automatic

Quads and Multiplexers

351

action as much as possible, and not require the operator to intervene.

12.9 INTEGRATED MULTIPLEXER AND DVR

The VCR has been replaced by the DVR in many video security systems. Consequently, many DVRs are now incorporating the video multiplexer into the DVR unit. The combined video multiplexer–digital recorder simpli-fies and reduces errors during the hardware setup pro-cedure and simplifies the design, operation, and cost of the system. Figure 12-10 shows a full-featured DVR– multiplexer combination. Table 12-2 lists some of the mul-tiplexer equipment available.

12.10 REMOTE DISTRIBUTED MULTIPLEXING

Digital technology is finding its way into the use of multiplexers in LAN, WAN, etc. as a superior technique for distributing, controlling, and recording video signals especially at remote distances (Figure 12-11).

Some multiplexers designed for larger physical security installations lend themselves to distributed multiplexing. This permits groups of cameras that are located in

DVR/MULTIPLEXER

CAMERA INPUTS

TRIPLEX

DIGITAL

HARD DRIVE

COMMUNICATION

VIDEO RECORDER

DISPLAY

(1–16 TYP)

MULTIPLEXER

(80, 160, 320 GB)

DRIVERS/PORTS

ELECTRONICS

SYSTEM FEATURES: • COMBINED DVR/TRIPLEX MULTIPLEXER.

· NUMBER OF CAMERA CHANNELS: 4, 9, 16

TRIPLEX OPERATION: SIMULTANEOUSLY VIEW LIVE AND PREVIOUSLY RECORDED VIDEO IMAGES WHILE CONTINUING TO RECORD AT THE SAME TIME, USING EITHER ONE OR TWO MONITORS.

· LIVE IMAGE RESOLUTION: 720 × 480 NTSC.

· RECORD RESOLUTION: 720 × 224 NTSC.

NETWORKABLE VIA ETHERNET (TCP/IP)

· MOTION DETECTION WITH CONFIGURABLE SENSITIVITY LEVELS.

· DIAL UP MODEM.

· REMOTE ALARM NOTIFICATION FROM MOTION DETECTION.

· PRE-ALARM VIDEO.

· EXPORT VIDEO VIA USB PORT.

FIGURE 12-10 DVR-multiplexer system

352 CCTV Surveillance

CAMERAS

CALL

MAIN

1

MONITOR

MONITOR

KEYBOARD

2

1

9

3

MODEM

4

IEEE-1394 (FIREWIRE)

USB

ALARM INPUT/

OUTPUT DEVICES

POTS

DOMES

ROUTER

MODE

LAN

16

16 CHANNEL TRIPLEX

INTERNET

MULTIPLEXER/DVR

HARD DRIVE

CD READ/WRITE

MONITORING STATIONS

FIGURE 12-11 Remote distributed multiplexing

physically distant locations to be connected to a slave mul-tiplexer. Several of these remote multiplexers are con-trolled from a single master multiplexer at the central location. The master unit communicates over the Internet and provides the video processing, recording, and display process signals. It commands each slave unit to deliver the required camera channels from the slave to the master, where they can be combined by one or several multiplexer recorder systems.

Multiplexing digitally compresses the images of each video frame and transmits them over the digital network to a DVR and onto digital monitors. The video images are compressed before transmission and later decompressed to display them on the monitor.

12.11 SUMMARY

The multiplexer and integrated multiplexer–recorder (DVR) have become an important part of the video

surveillance hardware. It is a powerful tool capable of combining many video images onto one multi-screen dis-play, thereby reducing the number of monitors required to view the cameras in the system. It can call up cam-era images showing motion in the scene. It also provides the capability to prioritize incoming alarm signals from external sensors with the VMD alarms. In an analog net-work, the multiplexer can send the video images and cam-era identification signals to a VCR or DVR for proper synchronization on recording and playback to the mul-tiplexer. In a digital network, it can transmit the com-pressed images over digital networks to monitors and recorders to remote locations for remote site monitor-ing. The simplex, duplex/full duplex, and triplex types are available to provide a multiplexer solution to most applications.

Chapter 13

Video Motion Detectors

CONTENTS

13.1 Overview

13.2 Background

13.3 Functional Operation

13.3.1 Surveillance

13.3.2 Detection Probability

13.3.3 Motion Assessment

13.3.4 Scene Lighting

13.3.5 Training Function

13.4 Analog Video Motion Detector (AVMD)

13.4.1 Technology

13.5 Digital Video Motion Detector (DVMD)

13.5.1 Mode of Operation

13.5.2 Technology

13.5.2.1 Programming the digital VMD

13.5.2.2 DVMD Setup Procedures

13.5.2.3 Sensitivity Settings

13.5.2.4 Motion Detection Sensitivity

13.5.3 Hardware

13.5.3.1 Normal Mode

13.5.3.2 Trace Mode

13.5.3.3 DVMD Graphic Site Display

Maps

13.5.4 Features

13.6 Guidelines, Pros and Cons

13.7 Summary

13.1 OVERVIEW

The method by which current security systems trigger secu-rity alarms can be divided into two classes. At one end of the spectrum there are systems that sense physical move-ment, such as simple contact switches and PIR sensors. While all these systems can be quite varied in the technol-ogy they use, the systems have one thing in common: they can only recognize movement. On the other hand, there

are visual detection systems ranging from guards posted at specific locations to camera systems with analog video motion detectors (AVMD) or digital video motion detec-tors (DVMD). The DVMDs use monitors, real-time and/or TL VCRs or DVRs to discern between allowable activities, breach of security or provide identification of individu-als, and give instructions to a guard on what a response should be.

Any video security system should include the following four ingredients: (1) surveillance, (2) detection, (3) assess-ment, and (4) response. The VMD can be a part of the system hardware to provide the surveillance, detection, and assessment, and provide accurate detailed and con-cise information to the guard force, allowing the force to respond optimally. As a free by-product, the VMD also makes available a training tool to practice and perfect the guard response philosophy. To achieve high detection probabilities in any moderate to large security system, the integrated video system must operate with an automated VMD detection system.

The recent availability of affordable DSP techniques has forever changed the security scenario and eliminated the shortcomings of the simple motion detectors and first generation AVMD detectors. In simple terms, advanced DSP technology has brought intelligence into the world of DVMD. DVMD systems combine visual video presentation of the motion detection with recording technology. Intelli-gent VMD systems go a step further by using sophisticated DSP algorithms so that motion detectors learn or adjust to a changing or new scene, virtually eliminating false alarms that were prevalent in the analog and simple first generation DVMD technologies. Intelligent DVMDs can be programmed to overlook small changes in the scene such as rain, dust, moving tree branches that often render traditional VMDs unusable.

The useful security information displayed on a video monitor often comes from motion within the scene—a

353

354 CCTV Surveillance

moving person, vehicle, object, or some activity involving motion. Irrespective of the number of security monitors, it is important to have an alarming device to alert the guard to motion or activity in a scene. Medium to large video installations generate many camera scenes that must ulti-mately be displayed on monitors, but it is difficult for a security guard to watch multiple monitors over long peri-ods of time. The video multiplexer goes a long way in reducing the number of monitors the guard must view and at the same time increases the operator’s ability to react to real threats, but it is the VMD that electronically analyzes and monitors camera images to detect changes (motion) that are judged to warrant an alarm. The VMD provides an electronic alternative to a guard sitting and staring at the monitors, and can notify the guard immediately of situa-tions requiring attention. VMD systems operate to detect changes in a specified area within the camera FOV. They do this by comparing the light levels of camera pixels from one video frame to the next, looking for changes considered significant. In the simpler, lower-cost AVMD systems, large areas in the incoming frame are compared with those of a previous reference frame. This type of sys-tem works reasonably well indoors, where there are few changes in the scene and where lighting is constant. Ana-log systems are, however, susceptible to false alarms caused by lighting changes, debris passing through the camera FOV, small animals, ripples on bodies of water, or camera vibration. They are therefore not recommended for most outdoor applications and instead the DVMD is used. The microprocessor DSP-based DVMD can analyze thousands of picture zones and operate with low false-alarm rates even under severe light-level changes. Most DVMDs, with the exception of those using the latest intelligent image processors and learning algorithms, are not suitable for PTZ applications.

Environment plays a major factor in choosing the DVMD for outdoor applications. The DVMD can toler-ate some camera vibration, but the camera should be mounted as securely as possible. The DVMD can also toler-ate light-level changes as might occur when a cloud passes in front of the sun, without causing a false alarm. Some DVMD systems can subtract out or ignore inherent scene motions such as waving flags, leaves, or trees, so that they will not be a source of false alarms. Some have the ability to selectively sensitize and desensitize certain portions of the scene in order to prevent false alarms. They desensi-tize parts of the scene where inherent motion and no real activity is expected, such as leaves rustling on trees. This reduces the chance of false alarms.

After a target has been detected and classified, the DVMD tracks that object within the site as the target moves from camera to camera. Systems are now available that can display images from remote locations showing targets in motion. The system can detect, classify, locate, and track objects within the FOV of the camera. The operator has a mapped display of the site, highlighted with icons of the

various types of targets (cars, personnel, gates, etc.), and can see an icon of the moving car or other target on the digitized site map. The path which vehicles take is synthe-sized in the monitor display of the FOVs of several cameras that the car had passed and traversed. Actual video scenes are available by clicking on the icon. These type systems are finding use in environments such as airports, seaports, and large installations.

13.2 BACKGROUND

In its most general sense, a motion detector is an ana-log device that responds to movement recognized as a specific type and rate of change within a defined moni-tored area of coverage. The original motion detectors were designed to detect motion or movement in a stable back-ground by means of PIR technology using pyroelectric detectors. These PIRs sensed gross changes in movement but provided very little intelligence as to the cause of the movement.

A video camera provided with appropriate VMD pro-cessing electronics can make the camera operate as an alarm sensor. The VMD processing electronics memorizes the instantaneous video picture, and then if some part of the picture changes by a prescribed amount, the system generates an alarm signal to alert a guard or activate a video recorder. The AVMD or DVMD is connected into the video system as shown in Figure 13-1. The figure shows an individual entering a room and the successive video frames showing the person walking through the facility. The VMD will detect the motion of the person, highlighting the per-son on the monitor screen, and/or also producing a visual or audible annunciation to the security officer.

Two VMD processing electronic types have been developed: the first-generation analog and the second-generation digital. The DVMD provides significantly more capability and reliability but costs more. Surveillance of any scene is achieved by the use of conventional video cameras and lenses positioned throughout the area of interest at locations that permit recognizing an intruder or movement within the camera FOV. Cameras should be positioned so they can view all activity and targets of inter-est. Figure 13-2 illustrates the VMD’s place in the video surveillance system.

In the 1980s, several DVMD systems became available. These were large, complex, and expensive units with elec-tronic memory and logic that dissected a video image into zones. Each zone represented an area in which motion could be monitored. By dividing the video image into hun-dreds of zones, the target could be localized in the scene and defined in size and motion, than it could in the orig-inal AVMD system. The light level of each zone likewise could be analyzed providing further intelligence about the scene. These systems were only affordable by large com-mercial institutions and government facilities. It was only

Video Motion Detectors

355

TO MONITOR,

VIDEO

ANALOG

CAMERA

OR

RECORDER

LENS

OR PRINTER

DIGITAL

VMD

CAMERA/LENS

FIELD OF VIEW(FOV)

G

G

G

SUCCESSIVE VIDEO FRAMES SHOW PERSON WALKING THROUGH FACILITY

G

G

G

G

G

T = 0

T = 1

T = 2

T = 3

T=4 SEC

FIGURE 13-1 Video motion detector (VMD) in the video security system

SCENE
LENS

VIDEO MOTION

MONITOR,

CAMERA

RECORDER

DETECTOR (VMD)

OR PRINTER

DECISION ELECTRONICS

VIDEO

VIDEO

THRESHOLD CRITERIA:

VISUAL

SCENE LEVEL

FROM

TARGET SIZE

CAMERA

AUDIBLE

TARGET SHAPE

TARGET SPEED

SWITCHED

NUMBER OF TARGETS

ALARM

SIGNAL

OUTPUT

PRESET MOTION THRESHOLD

(OPERATOR CONTROL)

FIGURE 13-2 Video motion detection system and detection parameters

356 CCTV Surveillance

into the mid-1990s that digital electronic costs were suffi-ciently reduced to make the present DVMD practical in security applications.

The evolution of the AVMD to the DVMD provided a significant step forward in identifying the source of an intrusion or movement in a video scene by providing more intelligence to the security operator. Early analog systems were limited to monochrome video cameras since color cameras were not in widespread use during the 1980s. The modest electronics in the AVMD limited their use to indoor applications as they could not deal with all the uncontrolled lighting, weather, and stray motion interfer-ences in an outside environment. The introduction of the CCD camera in the 1980s and low cost color cameras in the 1990s initiated the advent of a totally new technology in VMD. This, however, was not sufficient to make the AVMD a reliable product for indoor applications and especially not for outdoor applications. In the mid- to late 1990s, however, the introduction of the DVMD in conjunction with the CCD camera with DSP improved motion detec-tion significantly. Digital circuitry and availability of inex-pensive solid-state memory brought about the widespread use of DVMD. The DVMD has the ability to dissect the video image and analyze the scene on a pixel-by-pixel basis, thereby allowing sophisticated analysis of the motion in the scene. These new DVMDs have proven to be very reliable for alarm management, and provide automatic intrusion detection and automatic recording of intrusion events. They are used in open areas and relieve console guards of the tedious monitoring of empty hallways, rooms, parking lots, and parking garage levels that have no activity. The improvements in reliability through sensitivity adjustments and digital analysis of movement on a pixel level has given credibility to the idea that video surveillance systems can and should perform automatic motion detection without individual camera scenes requiring active monitoring by a console guard.

Advanced programming for “specific act recognition” is just beginning to emerge from development. This is motion detection that recognizes unique and complicated motions associated with undesirable act phenomena. Recognized acts can be the typical movement of shoplifters, acts of physical violence, or phenomena such as fire. Video smoke detection software programs are currently being marketed. Also emerging is the coupling of motion detection with alpha-numeric character and biometric recognition. This takes the form, for example, of spotting license plates or vehicle signage and processing the numbers or characters remotely. Some facial systems provide recognition of spe-cific faces in a crowd and are finding their way to market.

13.3 FUNCTIONAL OPERATION

Before any AVMD or DVMD can be applied to a partic-ular application, its location—indoor or outdoor—must

be considered. In an indoor application, the light-level changes are usually predictable or at least not very signif-icant. Successful VMD operation depends on recognizing light-level changes in specific parts of the scene (caused by an intrusion or disturbance) in contrast to overall scene light-level changes caused by changing lighting conditions. These two phenomena must be differentiated to avoid undue false alarms. In indoor lighting applications where the light level is controlled by the user, a simpler AVMD system can be used.

13.3.1 Surveillance

Video surveillance is accomplished via the use of cameras and lenses located and positioned for maximum intelli-gence gathering of a viewing area. The cameras can act synergistically with other alarms as remote eyes to present a visual image of an area as well as the source for an alarm input.

Monitoring a large area such as a parking lot using VMDs presents multiple possibilities including: (1) a wide-angle lens, (2) multiple cameras, and (3) dual-lens, split-screen. When a wide-angle lens is used, the alarm source (intruder) appears small on the monitor screen and a guard does not detect the intruder, especially if the intruder takes cover quickly. The VMD can detect the intruder and register an alarm. With multiple cameras, the parking lot FOV is divided among the cameras, each viewing a section of the overall area. Each camera must use a separate VMD. With the split-screen technique one lens can be wide-angle, the other a medium or narrow-angle lens.

If the system includes pan/tilt equipment the guard must pan, tilt, and zoom the camera/lens to locate the alarm source. This is not a simple task, and in the time required for the guard to perform it, the intruder may be gone. In more sophisticated systems, in order to speed reaction time, the location of the motion in the image is used to point the pan/tilt platform in the direction of the motion.

13.3.2 Detection Probability

The protection of outdoor areas presents the most difficult problem in facility security. All sensing devices are plagued by false alarms due to the unpredictable nature of natu-ral phenomena and intentional artificial alarms. Seismic sensors produce false alarms due to vibrations caused by wind, vehicles, and other objects. Microwave sensors pro-duce false alarms due to moving animals, blowing papers, or leaves. An effective outdoor security system is best aug-mented using video cameras viewing the actual scenes to filter out and recognize false alarms. Although an alarm denotes that a certain area has been disturbed, without a

visual image little information is provided as to the nature of the alarm or the precise location at which it occurred. Without a video image, security personnel must be sent out to investigate and determine the nature of an alarm. Since outdoor monitored areas are often large, in many cases by the time a security guard responds to the alarm, the intruder is gone or the activity has ceased.

A guard monitoring a medium to large video secu-rity system must view many monitors that display either:
· sequenced scenes, (2) several monitors—one for each camera, or (3) monitors with split-screens. To assure a high probability of detection, the camera lens magnifica-tion must be such that an intruder is displayed on the monitor magnified enough so that the guard can easily see him and attract his attention. Using multiple cameras is often the best solution to provide the necessary coverage to detect the intruder.

For a guard’s response to an intrusion to be effective, the guard must first know that he is responding to a real intrusion, its location, and nature. The VMD function is to display only intrusion alarms on the video monitor with-out any human intervention. The guard then assesses the alarm by viewing the monitor.
The VMD system must give timely information as to the exact location and nature of the activity and must:
· respond to small changes (motion) in the camera/lens FOV, (2) activate an alarm output on the monitor to alert the guard that an intrusion has occurred, and (3) dis-play the alarmed scene on the monitor. It should also be accompanied by an audible and/or video alarm and acti-vate a VCR or DVR and video printer. For larger digital infrastructures it should be able to provide transmission of the video image over a network. The displayed scene should show the location within the scene that has been activated and give immediate information to security per-sonnel as to the precise location, movement, and nature of the alarm. If an intruder is hiding, a flashing pattern on the monitor should show the path of the intruder from entry of the scene to the point to where he is hiding.

Intrusion detection probability is controlled by the placement of cameras and is a system design parameter. The ideal motion detection system would give a 100% probability of detection of intrusions, zero false-alarm rate, zero nuisance alarms, and zero equipment failure. With proper camera placement and reliable equipment, target-detection probabilities can be 95–99%. Alarm assessment takes place in the time it takes for the operator to view the scene and identify the cause. When a VMD is used, the security operator does not have to identify the camera or locate the movement on the screen, since the cause of the alarm is indicated by the brightened flashing map on the monitor. If it is an intruder, the guard responds accordingly, knowing where the intruder is and who he will be confronting. If it is not an alarm, the guard can press an alarm reset button and go on to the next alarm.

Video Motion Detectors

357

Video motion detectors are valuable not only because they can cue a video response but also because they are an independent source of vital information. There may be particular situations where a specific activity within an area covered by the camera would be difficult to detect with other conventional forms of alarms. It is often important to know not only that an intrusion occurred in a certain space or area but also the path the intruder took. VMDs with enhanced mapping display capability can provide this information.

13.3.3 Motion Assessment

Assessment is the ability of the console operator to identify and evaluate the cause of the alarm. This judgment call is one of the most important decisions for two reasons:

· if a real intrusion occurs the guard’s assessment must be rapid and accurate and depend on a visual judgment,

· if the alarm is not a valid intrusion, the guard must be able to make that decision rapidly and accurately— which again requires visual observation of the cause of the alarm—and then cancel it.

In some DVMD systems a RAM module stores the alarmed locations in a separate RAM alarm map mem-ory (AMM). Upon alarm, the contents of the AMM are displayed on the alarmed video monitor scene as a flash-ing, highlighted array of alarm points. This feature is a key to quick, accurate assessment of all alarms. The AMM enables the operator to determine instantly the exact loca-tion where the disturbance or intrusion has occurred and provides a quick, precise evaluation of the alarm to provide the appropriate response. To clear the alarm condition after a response has been made, the operator presses an alarm reset switch and the monitor returns to the normal blank condition. This accurate, rapid assessment optimizes the use of the response force. If a second or additional alarm occurs prior to resetting, the alarm scenes are dis-played with their alarm maps in sequence on the master monitor, at a selectable rate.

When a large number of cameras are alarmed simulta-neously, an assessment problem can occur. By the time the guard views the last camera, the intruder most likely has left the scene and only the map remains. The DVMD effectively controls the situation by providing a video out-put to record all alarmed camera images. This is done automatically while the guard watches the monitor. The video frames (scenes) are sent to the VCR or DVR at a rate of 30 fps. The pictures are recorded—one from each camera—in sequence and continue until the operator resets the equipment. When a guard realizes a multiple-intrusion attempt is in progress, the guard can playback the recorded video images into the monitor and replay the intrusion with the alarm map to determine the cause of the alarm in the scene. Using this technique the alarm assessment capability is extremely high. The guard need

358 CCTV Surveillance

not leave the console during an alarm condition unless it is necessary to initiate a direct response to a real intrusion. The guard can observe the progress of the intruder into the area by observing the monitor as the intrusion map is generated.

optimally. This important training improves the plan, the guard response time and method, and overall security.

13.4 ANALOG VIDEO MOTION DETECTOR (AVMD)

13.3.4 Scene Lighting

Since the VMD makes its decision based on the scene the camera is viewing, it is important that lighting at the camera site is adequate. The VMD equipment must be able to compensate for variations in average scene lighting occurring during daylight hours as well as when auxiliary artificial lighting is provided during nighttime operation. VMD systems operate with scenes illuminated by visible or infrared lighting.

In outdoor applications, the environment is not as con-trollable: significant light-level changes are caused by sun-light, cloud variations, lightning, and many different types of objects passing through the camera/lens FOV. Many DVMD systems operate well under most outdoor condi-tions but they lose some of their capability under adverse environmental conditions of heavy snow or rain, and alter-native systems using other sensors should be relied upon. The DVMD used in an outdoor environment has a signifi-cantly higher potential for false alarms due to these unpre-dictable lighting changes and moving clutter. The DVMD must have outdoor algorithms that correctly account for these rapid changes in overall scene brightness and illumi-nation, as well as area changes in illumination caused by rapidly moving phenomena. If there is movement in the scene it must be detected while the movement is still in the scene. Therefore, if updates of the scene occur at too slow a rate, an object at a distance may elude detection.

To determine whether a target is of interest or a false alarm, the equipment must be able to distinguish its size, speed, and shape. In outdoor applications a DVMD is the only solution.

13.3.5 Training Function

In the intrusion scenario, when an alarm occurs the con-sole operator is called upon for the first time to evaluate the alarm on a previously blank CCTV monitor. The mon-itor displays the intruder and the exact location within the scene by some flashing indicator superimposed on his exact location.

Management uses AVMD, DVMD, and video recorders to test a security plan and guard response, and evaluate guard and overall system performance. A system using the motion detector permits security personnel to train before an actual event, and when an intrusion does occur, the sys-tem can immediately recall the decisions to form an instant plan of action. This directs the efforts of the response force

For several decades, the AVMD has attempted to identify motion and activity of interest in a video scene. It has enjoyed some degree of success for indoor applications but has not been successful in outdoor environments. With the recent introduction of the DVMD in conjunction with DVRs and digital multiplexers, VMD has now become an important, even essential, tool for video monitoring.

The AVMD system is simple: it monitors any change in the video signal that comes from the camera and produces an output indicating that there was an alarm. Unfortu-nately, many other changes in light levels are not caused by targets of interest but rather from background changes. The particular causes for these false alarms are:

· An overall change of the scene lighting caused by sud-den light changes or fluctuations in overall lighting, and turning lights on and off

· Flashing a light across a scene causing an immediate contrast change

· Open flames, flashing neon signs, cigarette lighters

· The sun passing behind a cloud

· Flying debris: flying paper boxes, etc. through the cam-era FOV

· Environmental dust, a rainstorm, or snowstorm

· Animals, birds passing through the camera FOV

· Continuous motion from water fountains, revolving doors, escalators, ripples on water, or wave motion.

For all these reasons, the AVMD is not a viable solution for detecting motion, real target, or activity in a video system, and does not find widespread use except in small systems.

13.4.1 Technology

The AVMDs have been available for many years and pro-vide a low-cost video device to detect simple motion in a video scene. They operate reliably only in indoor, well-controlled environmental and lighting conditions and should not be used for outdoor applications. Figure 13-3 shows a block diagram of the AVMD.

The simplest AVMD uses analog subtraction. The refer-ence frame and the frame in which motion has occurred are subtracted and an alarm declared depending on the amount of signal difference between frames. This analog system, while acceptable for most indoor applications, is prone to false alarms and is not suitable for outdoor appli-cations. A digital DVMD should be used in all outdoor applications.

Video Motion Detectors

359

SECURITY

OPERATOR

CONTROLS

RESET

VIDEO OUTPUT

VIDEO

MONITOR

SENSITIVITY

OUTPUT

RECORDER

INTERFACE

PRINTER

CAMERA

WINDOW

ALARM OUTPUT

VIDEO

SIGNAL

AUDIBLE ALARM

SIGNAL

SIZE,

SHAPE,

LEVEL

VISUAL ALARM

CONDITIONING

LOCATION,

COMPARATOR

LOGGING PRINTER

GENERATOR

REFERENCE:

AUTOMATIC

ADJUSTMENT OF

SLOW CHANGES

IN LIGHT LEVELS

FIGURE 13-3 Analog video motion detector (AVMD) block diagram

Two generic detection options available in many VMDs are: (1) detection of motion or activity, (2) detection of the presence or absence of an object. These systems can be configured so that these two different type windows operate independently and be can be combined within the same camera FOV. Motion windows are designed to detect movement of objects or personnel into and through their detection zones. They also detect anything that moves into the window and stays there even though the object stops moving. They can have a programmable time-out feature so that an object can enter the detection window and stay there for a given length of time without causing an alarm. This ensures that the DVMD does not indefinitely remain in an alarm mode. The motion windows look for significant changes in image contrast or pattern in the detection zone. They detect only significant changes in most objects that are bright or dark but are much smaller then those expected from some debris, and will not trigger a false alarm.

In the object presence or absence mode of operation, the system monitor displays the movement of objects that are expected to remain stationary during the surveillance while ignoring surrounding movement. If particular assets are to be protected and can be defined in space, the VMD defines a tight window around the object to instruct the system to signal an alarm if the object moves while ignor-ing anyone passing through the FOV. When using either of the two modes the individual windows are augmented by background scene monitoring functions so that the overall scene illumination levels are monitored to detect and compensate for sudden light level changes.

All AVMDs have an adjustable detection-of-motion zone (DMZ), which is a selected portion of the monitor screen. Any movement (change of light level) in the scene within the DMZ automatically triggers any one of four alarms:

· an internal audible alarm, (2) a front-panel signal light, (3) an AC or DC outlet that can activate an AC- or DC-operated signaling device, or (4) an isolated terminal relay contact to activate a video recorder, printer, bell, or other security device.

On most AVMD equipment, the size, shape, and loca-tion of the active area in the entire scene is adjusted with front-panel controls. The DMZ size and configuration chosen depends on the requirements of the surveillance application. Figure 13-4 illustrates some examples of DMZ shapes available, including split-screen, square, rectangle, L-, C-, and U-shaped.

The areas of sensitivity are chosen to surround a location in the scene where motion is expected. The DMZ enables the operator to select (sensitize) specific portions of the camera scene area, while the entire scene is always dis-played. An alarm occurs only if there is motion in the DMZ itself. Depending on the equipment, DMZ is represented on the video monitor screen by a brightness-enhanced window (or a brightness-enhanced frame), adjustable via the front-panel controls. After initial setup, the brightened window (or frame) may be switched off so that the scene looks normal to the operator. The active DMZ on the screen can be set up to cover an area anywhere from 5 to 90% of the viewed picture width and height. The AVMD system sensitivity is usually set to respond to a 25% change

360 CCTV Surveillance

MONITOR SCREEN DISPLAY

SPLIT SCREEN SQUARE RECTANGULAR

L SHAPED C SHAPED U SHAPED

VMD SENSITIVE TO ALARMS IN CROSSHATCHED

AREAS ONLY

FIGURE 13-4 Detection of motion zones (DMZ) in analog video motion detectors (AVMD)

in video signal level, in 1% of the picture area occurring within a time period of several frames.

The AVMD operates by analyzing the analog video sig-nal from the camera and determining whether the scene has changed. The system “memorizes” the value of a stan-dard reference scene depicted within the DMZ and com-pares it with a value in the current real-time scene. If the two values are the same within the active DMZ, electronic circuitry declares that there has been no motion and no alarm is declared. On the other hand, if there has been a scene change caused by someone intruding into the scene, an object moving, or some other light-level disturbance, providing the change is larger than a prescribed amount, typically 10–25%, then electronic circuitry decides that a change has occurred, there has been motion in the alarmed area, and an alarm signal is produced. This alarm signal is used to produce an audible or visual alarm, turn on or activate a video printer. The AVMD operates inde-pendently of the video monitor or any other recording equipment, and in no way interferes with it.

13.5 DIGITAL VIDEO MOTION DETECTOR (DVMD)

While analog VMDs have been in use for security appli-cations for many years, they have only been moderately

successful in indoor applications where lighting has been well controlled. In outdoor applications, a far more com-plex digital electronic system is needed to provide reli-able VMD capability. The DVMD must take into account the many variations of lighting, type of target movement, and electrical background disturbances caused by exter-nal sources and noise in the system. In the past, these sophisticated expensive systems have been used in large government facilities and nuclear power plants. With lower cost derived from high density memory and more power-ful computers, the DVMD is now in more widespread use in commercial installations.

The DVMD allows the user to divide the monitor’s video scene into small detection areas called windows, and in some cases even smaller size areas going down to the pixel level. The flexibility of these windows allows the user to specify particular areas or zones of interest. Each window or zone has its own set of programming levels for sensitivity and alarm triggering level. Only the windows are activated or processed for alarm events: all the other parts of the scene which either are not of interest or may contain false alarm producing motion are not. Using this technique, doors to a building may be monitored while headlights from an adjacent car parked or other bright lights in the scene are ignored. Since average light-level changes in the scene occur, the system automatically adjusts to both increasing and decreasing illumination by monitoring and

updating reference levels for each video input. The entire scene is also continually monitored for light and illumina-tion changes and full image scene changes such as those caused by a lightning or clouds drifting in front of the sun. The scene changes would not trigger an alarm but rather reset the references for each window, and the VMD would continue monitoring the detection zones for motion or inactivity. The sensitivity of each window is monitored and controlled by the user.

The more sophisticated and expensive DVMD systems use elemental detection zones, in which the scene is divided into a large number of zones (hundreds to thousands) and converted into a digital signal. The processor analyzes these individual zones and makes a decision whether or not an alarm is present. With these microprocessor-based systems, many parameters are ana-lyzed, thereby forming a more reliable basis for an alarm signal decision. Light-level changes in these DVMD sys-tems are compared with the previously stored values ratio-metrically—that is, on a percentage basis. Ratio-metric thresholding causes the system to cancel out any gross change in the scene lighting, so that an alarm decision is made strictly on an incremental basis, for a small portion of the total picture area.

The digital electronics in the DVMD subdivides the cam-era scene into many small elemental zones—as many as 10,000—and makes a zone-by-zone comparison (subtrac-tion) of the non-moving or steady scene with the motion scene. It goes into an alarm mode when a threshold is detected in any one or a multiple of these zones. By con-verting the signal from analog to digital and dividing it

Video Motion Detectors

361

into many zones, a much more sensitive device results. This technique allows discrimination between real targets and false alarms and other scene lighting variations, and provides a more reliable system for outdoor use.

The user-selected zones are positioned over specific areas where motion is expected. These zones may cover assets to be protected, entry or exit points, parking lot slots, perimeter areas, and perimeter fence lines. Each zone may be set with a different sensitivity appropriate to the per-centage change required to trigger the alarm in that zone. The larger the percentage required to cause an alarm, the less sensitive the system is to contrast changes and the less likely it is to produce false alarms. The DVMD is much more sensitive than the large area detection AVMD.

13.5.1 Mode of Operation

The DVMD processing unit converts an analog video signal into a digital code and performs DSP to make it sensitive to specific types of motion in the camera scene (Figure 13-5). For each camera a specific detection pattern or area is selected, or already programmed into the electronics memory. The detection pattern is part or all of the camera image scene within which specific sample points are desig-nated. Depending on the manufacturer, the sample points vary in number and location. At a designated rate, the sam-ple or reference image from a specific camera is converted from the analog to the digital format, and the digital val-ues are stored in temporary memory in the VMD unit. This reference or base image is updated at variable rates

OPERATOR CONTROLS:

SCENE LEVEL

TARGET SIZE

TARGET SHAPE

TARGET SPEED

NUMBER OF TARGETS

DIGITIZED

CELL

VIDEO OUTPUT

ANALOG

VIDEO

(ZONE)

VIDEO

(BY PIXEL)

VIDEO

MONITOR

CAMERA

VIDEO

ANALOG TO

TRACKING

RECORDER

DIGITAL

PREPROCESSOR

PRINTER

VIDEO IN

CONDITIONING

PROCESSOR

CONVERSION

VIDEO

ALARM

OUTPUT

ALARM/CELL BOUNDARIES

CELL/ALARM

DEFINITION

VIDEO AND ALARM/CELL
ANNOTATION (TO MONITOR) SECURITY OPERATOR

INPUT

FIGURE 13-5 Digital video motion detector (DVMD) block diagram

362

CCTV Surveillance

to compensate for small changes in the scene that do not

target is viewed from a distance it appears to have a small

constitute alarm events.

size on the monitor image. As the target moves closer to

At programmable rates at a later time, the camera

the camera it increases its apparent size thereby causing

images are converted into a digital format and electroni-

the confusion in target identification. Motion detectors

cally compared with the stored reference image. If there

generally have a more positive identification of a target

has been movement in the scene or any variation in a

if the target is moving perpendicularly or at an angle to

significant number of sample points over some range, an

the camera, rather than toward or away from the cam-

alarm is triggered. If some harmless objects such as a

era. If cameras can be mounted to have this relation-

small animal or bird or debris pass through the scene no

ship to the target, a positive identification can usually

alarm will occur. If, however, there is movement within the

be made. The most significant new parameters added to

scene—such as a person entering a window or opening or

digital VMD processors to improve the capability for out-

closing a door—the VMD will be triggered. The number

door operation have been: (1) improved multi-directional

of sample points and the amount of change within the

detection, (2) 3-dimensional perspective analysis, and (3)

areas to produce an alarm output depend on the particu-

automatic adjustment to changing environmental condi-

lar manufacturer, model, and operator control settings.

tions. Improved multi-directional detection provides the

Depending on the design, a VMD can process 1, 10, 16,

ability to determine whether the object is moving directly

32, or 64 cameras and sample them serially: that is, camera

toward or away from the camera, especially when the tar-

1, then camera 2, and so on, and then back to camera 1.

get is at a distance. The ability to automatically adjust to

Some systems sample and process multiple cameras simul-

changing environmental conditions removes the technical

taneously, then analyze and respond to multiple alarms.

difficulty to manually readjust the system sensitivity set-

When a VMD detects an alarm event its output can be used

ting to match daily weather variations. Systems not having

for multiple functions. It can display the alarmed camera

this ability are difficult to calibrate and require constant

on a monitor, alert a guard with a visible or audible sig-

recalibration.

nal, record the alarm on a video recorder, send the alarm

signal to a remote site, or activate a TL VCR or DVR with

an alarm input to change its recording mode from TL to

13.5.2 Technology

real-time.

In contrast to the AVMD that detects the change in light

When a video image is converted to data in a digital for-

level in one or a small number of scene locations (zones),

mat, the image information becomes the stored digital

the DVMD electronically analyzes hundreds or thousands

value. This digital value changes as the video image (the

of zones in the video signal and provides information such

source of the data) changes. Complex algorithms analyze

as the location in the picture where a motion or intrusion

has occurred. Its output drives various audible and visible

the changing digital values to recognize patterns. This is

alarm signals, a graphic monitor map showing the motion

considered as video content analysis. These algorithms are

path in the image, and a record of the intrusion using

a software function and are programmed into electronic

a recorder or video printer. In normal operation when

chips that can be installed in cameras, standalone mod-

there is no motion or change in a scene, the VMD takes

ules, DVRs, and dedicated computer processors. DVMD

the video signal from the camera, stores the video frame

is also available as software for installation in off-the-shelf

(containing no motion), continually updates and memo-

computers.

rizes the subsequent frames, and compares them to the

Algorithms have been designed to decrease the number

previous frame to see if there is a difference in the new

of monitors that must be viewed. This is accomplished

frame. If there is no motion there is no alarm. If there

by scene averaging and filtering techniques to eliminate

is a difference of measurable and defined value, then an

items that do not fit the model of the motion or activity

alarm is declared and an output produced.

and do not represent a threat to the site. Once the system

Caution must still be taken for outdoor applications,

detects an object, it applies various tests in an attempt to

however, in which there are rapid changes in sunlight,

classify the object, taking into account such characteristics

clouds, shadows, distance of objects, rain, snow, movement

as size, shape, true height to width ratio, and location. If

of trees or shrubbery, camera movement in winds, automo-

the object or activity fits one of the criteria for a target, it

bile lights, ripples on the water, and other small moving

is marked and a more accurate determination is made to

objects. This can represent a fairly impressive range of

identify personnel and activities.

problems that must still be considered in outdoor appli-

Digital VMD technology has the ability to monitor every

cations. To address some of these problems, DVMD sys-

pixel of every image individually and/or as a group. The

tems have additional automatic adjustments (algorithms)

light level of each pixel can be memorized in storage and

to process the visual signal data to exclude some of these

compared to subsequent images to determine if there is a

problematic false alarms. One problem, in particular, is

light-level change and how much the change is. By apply-

to determine the size of a target in the scene. When a

ing this technology over the entire image, the light-level

changes in each pixel can be examined and a determi-nation made whether it fits the criterion of an alarm. Algorithms are designed to identify objects of specific size, shape, movement, etc. on a pixel-by-pixel basis. Flying debris and other false alarms can be filtered out by size, object direction and speed, color, and type of motion and pattern.

Determining the size of an object in the FOV is difficult since the object appears as a different size depending on its distance from the camera. If the object is close to the camera it is large and as it moves away from the camera it becomes smaller and smaller. For this reason, parameters such as shape and movement are also required to deter-mine the identity of the object. Object direction can be determined easily since the object activates many pixels and by keeping track of the left-to-right or up-and-down motion it is easily accomplished.

In some cases the color of the object may be useful, and this is easily determined in the color camera by mon-itoring the color of each pixel in the moving object. This can be important if a person with particular color cloth-ing has been identified as the target. The parameter of color is used to continue tracking that person. Likewise, in outdoor applications if an automobile is identified with a particular color, the color might be the most important criterion for tracking the vehicle. Environmental condi-tions producing dust, fog, rain, snow, and sleet produce some ambiguity in target detection. These disturbances generally reduce the range over which VMD is effective.

Combining object motion and pattern recognition can provide additional information in determining the identity of a person and the behavior of the target. Algorithms have been devised to identify the movement of a person walk-ing. They have been able to tell the difference between a person walking, a walking dog, a crawling man, and oth-ers. There are also various motions that an intruder or criminal makes as compared to our normal movement, and these abnormal motions can be saved and put into storage and can help to identify a person exhibiting such movements in the video image.

An object’s speed is used by setting criteria for how fast the object of interest is able to move, and if the object is moving faster or slower than a predetermined speed it is registered as a false alarm. The VMD can have a library that stores information about the unique movement and pattern of particular objects such as paper leaves, ripples on water, birds.

The DVMD has the ability to remove constant motion from the scene which often takes the form of rain storms, snow, sleet, hail, water fountains, waves on water, etc. Algo-rithms stored in memory are used to filter out these con-stant motion disturbances. If there is an object within such constant motion moving at a different speed the system is able to identify this target.

The DVMD digitizes the frames from each camera into a large number of zones corresponding to exact locations on

Video Motion Detectors

363

the monitor screen. The number of digitized zones varies from hundreds to many thousands. The system assigns an absolute gray-scale value (light level) to each zone and stores the digitized gray-scale value and location in RAM. This procedure is carried out for each video camera chan-nel. The DVMD can digitize the picture into 16–256 gray-scale levels, thereby storing (memorizing) the image scene very accurately. After this reference scene has been mem-orized in RAM, the DVMD digitizes subsequent camera frames and compares them to the stored values, zone by zone. If the stored levels at any location differ by one or two gray scale levels—between the stored frame and the live frame—an alarm condition exists.

Most DVMDs in use today use standard menu screens to monitor and respond to alarms, using either simple keyboards or a mouse device for programming, adjust-ment, and normal operation. Most current systems do not require a personal computer (PC) for operation, but all provide an RS-232 interface for computer integra-tion or remote programming and reporting. The RS-232 approach and menu-driven screens for operation and con-trol of the digital VMD systems provide a friendly interface to the user.

Self-contained DVMDs are based on proprietary signal processing algorithms and easily integrate into existing multi-camera video systems. Most camera inputs are digi-tally sampled with a resolution of 768 by 480 pixels and eight bits (256 levels) of grayscale. All images are sampled and displayed at 30 fps (60 fields per second). Each cam-era is associated with a dedicated event when an alarm output occurs, and can be connected to a video recorder or audible or visual anunciator whenever any window in any camera has been alarmed. Additionally, a video loss output signals an alarm if the camera loses power or no video signal is present, and remains active until the video signal is restored or the time-out feature resets.

Many DVMD systems have two monitor outputs although only one monitor is required for viewing. Many users prefer a dual-monitor approach. One monitor is used to view live sequencing from camera to camera or a specific camera view. The second monitor is used in dig-ital mode to view motion detection windows triggered by an alarm. When alarms occur from multiple cameras, the operator can sequence through the alarming cameras at a user-defined rate or go to the quad or 9 or 16 split image display with the alarming cameras in that mode. In any case, the images from the alarm cameras are highlighted graphically on the display. The minimum hold time for each alarm is user-defined, usually from several seconds to 5–10 minutes. The user can also select freeze times for any of the alarmed images ranging from seconds to minutes. In the freeze frame mode the video display is locked into a full screen. When an event occurs in that camera after the freeze frame time has elapsed, the video continues in full motion allowing the guard to continue monitoring the cameras.

364 CCTV Surveillance

In the playback of recorded images from the VCR or DVR the output can be displayed on either or both mon-itors. This allows one monitor to be left in the normal display mode monitoring potential alarms while the other plays back the recorded images for review.

13.5.2.1 Programming the digital VMD

The DVMD includes an RS-232 interface to allow the user a choice of using either the front-panel controls or a mouse for system setup. Either way the window placement, size, or sensitivity are simply defined. Each camera can be programmed to include numbers and titles defining the specific camera, which is later displayed whenever that camera is displayed. These titles may be positioned anywhere within the full screen window so as not to obscure any important areas in the image. The system utilizes pull-down programming menus to control split-screen sequence rates, the camera ID information, and any other titles. Menus are available to adjust the sensi-tivity and scene area balance of the pixel level for alarm functioning.

Some systems can provide not only intruder detection but also lost object detection. Even in the presence of multiple moving objects in the same window, intelligent DVMD systems can accommodate a rapidly changing illu-mination condition commonly found in outdoor scenes, as well as sudden illumination changes from man-made and natural sources.

13.5.2.2 DVMD Setup Procedures

The DVMD system uses graphic symbols for motion sen-sitivity settings, simplifying the motion detection setup. In addition to a flashing cursor on screen, text prompts appear as shown in Figure 13-6.
Cameras can have motion detection in particular areas in the scene completely disabled. This should not be confused with enabling or disabling individual zones or pixels in areas of interest. Disabled zones that may con-tain unimportant or incidental movement include the following:

· Trees that can sway in the wind

· Pedestrians and vehicular motion that is not important

· Reflections from glass, bodies of water, or other highly polished surfaces, which can be sources of apparent motion.

The different alarm zones can be designated on the monitor in different colors for identification purposes. Examples are:

Choice Color of Flashing Cursor

No action Gray/white

Enable zones Black/white

Disable zones Clear/white

ACTIVE ZONE SETUP

ALARM AREA OF ACTIVITY

DISABLE PROBLEM ZONES

PEDESTRIAN AND VEHICLE

MOTION THAT IS NOT

IMPORTANT
MOVING TREES, BUSHES
CLOUDS IN SKY
RIPPLES ON WATER
REFLECTIONS FROM GLASS
OTHER

MOTION

SENSITIVITY

GRAPH

FIGURE 13-6 On-screen digital video motion detector (DVMD) graphic display

13.5.2.3 Sensitivity Settings

A bar graph is often used to illustrate the alarm sensitivity setting for the camera. The bar graph displays the sensi-tivity setting as a red line. A black line moves from the bottom to the top of the bar to indicate a change in motion or activity in the scene. When the black line reaches the red line above, a motion alarm is activated (Figure 13-7). The user selects a number or sensitivity button between 1 through 10 to change the sensitivity. In practice, watching the scene from a camera and watching the motion helps to determine the appropriate sensitivity setting for the cam-era. This procedure is performed for each camera during the initial setup phase of the system.

13.5.2.4 Motion Detection Sensitivity

Motion detection sensitivity for each camera can be set to levels from 1 through 10. The setting is made on a camera-by-camera basis, and applies to all enabled zones in any particular camera scene. Each of the zones distinguishes among 256 grayscale levels averaged over each zone’s area. A sensitivity of 1 is the least sensitive to motion and a set-ting of 10 is the most sensitive to motion. These settings are made using a bar graph similar to that used in the sen-sitivity settings above. Some recommendations for setup are listed below:

· If motion detection activates without an apparent cause, reduce the sensitivity.

Video Motion Detectors

365

· When setting sensitivity, select the highest setting that does not result in frequent false motion detection.

· The higher the sensitivity, the more likely the incidental movement to be detected as motion.

· When setting high sensitivity, such as 8–10, sources of false motion like reflections and windblown trees should be absent, otherwise alarms will occur.

The DVMD used as a sensor activates alarm inputs, essen-tially creating a motion-based alarm sensor input. The system in this scenario does not distinguish between an input from an external alarm sensor (switch, PIR, glass break detector, etc.), or when activated internally to the VMD system.

13.5.3 Hardware

Some DVMDs monitor up to 32 separate video cameras by sampling, time-sharing each camera sequentially. Each camera can have a separately adjustable sensitized alarm-ing area, thereby optimizing each camera to the scene it views. Likewise, the number of sensitive zones in each cam-era is chosen independently to match the scene require-ment. If one camera views a large area scene looking for small intrusions, the operator can make the alarming zone small for this first channel. If another camera views a small area scene looking for large intrusions, the operator can make the alarming zone large for this channel, and so on. Equipment setup procedures differ from manufacturer to

ACTIVE ZONE SETUP

AREA AROUND THE HOUSE ALARMED

GRAPH CHANGE INDICATES DETECTION

MOTION

SENSITIVITY

INDICATOR

PEDESTRIAN

MOTION

DETECTED

IN ONE CELL

FIGURE 13-7 Bar graph sensitivity display

366 CCTV Surveillance

manufacturer, but there are some common parameters and controls that must be determined and set when ini-tially installing the DVMD system. Typical setup controls include:

· Channel Mode Control. A switch selects the mode for each video camera channel. In the down position— INHIBIT—the channel is disabled and no alarms are registered. In the middle position—NORMAL—the cameras are ready for motion detection and alarming. In the up position—SET—the console operator can man-ually select any camera on the alarm monitor. When released from the SET position the switch returns to the NORMAL mode.

· Alarm Area Control. The alarm area control lets the operator manually adjust the position and size of the alarmed area zone. These adjustments can desensitize areas of the camera’s FOV where normal movement would cause an unnecessary alarm. For example, in an outdoor scene where a flag is constantly waving, the desensitized area would appear on the monitor but movement within that area would not cause an alarm.

· Refresh Control. The refresh rate refers to the time interval during which the reference frame memorized in RAM is stored, before it is again updated. Systems use refresh rates varying from 1/30 second up to several seconds. The operator selects the refresh rate, which is normally a function of the number of cameras and the kinds of alarms expected in the scenes.

· Ranging Control. Most systems allow adjustment of the electronic analog dynamic range of the analog-to-digital (A/D) converter. The function of the A/D converter is to change the camera’s analog electronic video signal to digital values. To provide the best scene resolution for each camera, the operator adjusts the range of white to black level in the digitized video signal.

· Masking Control. The masking control allows the oper-ator to enter scene areas on the monitor screen for which no alarming will occur. It is entered by inserting rectangular, square, or other masked areas. In some sys-tems the operator enters the masking with a light pen. The light pen permits irregular shapes to be desensi-tized merely by drawing around the object in the CCTV monitor scene.

In many VMD systems the detection zones may be of any shape and be divided into separate areas to accommodate unique detection requirements. Zones can be individu-ally turned on or off to accommodate entrance, hallway, parking area, or other locations. Two examples of zones being turned on or off individually are the following: (1) a zone encompassing a gate or doorway can be turned off during shift changes while other zones in the same scene can remain active to alarm and alert an operator of unauthorized intrusions, (2) a zone encompassing a file cabinet can be left off during normal working hours and turned on overnight. The systems can have independent

16-step zone sensitivity, signal integration (retention), plus multilevel digital filtering to maximize motion alarm detection and minimize false alarms. Periodic automatic rebalancing minimizes the effect of slow light changes, such as those occurring between daylight and nighttime conditions.

In operation, a cell is activated by the changes in the video content of successive picture fields. A higher retention setting delays the automatic rebalancing to opti-mize detection of slow changes or slow-moving objects. Both the video change (sensitivity) and the rebalancing time (retention) assigned to a zone can be adjusted to opti-mize detection and minimize false alarms for that zone. Any activated cell in a zone alerts (activates) that zone and channel.

Systems have integral video switchers with dual video outputs and RS-232 port to allow the DVMD to function as a standalone system. An audio output is available to warn the operator of an alert, and a relay closure can start a recorder for recording alerted channels. The RS-232 ports provide both a control input and an alarm output. They permit remote system control via a separate control keyboard, a data terminal, or a computer.

Either of the two on-screen alert presentation modes may be selected to highlight the intruder’s path through the facility. They are normal or trace.

13.5.3.1 Normal Mode

In the normal mode, a bright dot is displayed in the pic-ture on the alarm monitor at the center of each activated cell. With manual reset, this dot remains lighted until the channel is reset. With automatic reset, each dot disap-pears 16 seconds after the cell was first activated. Thus an intruder moving into a zone will cause a series of dots to appear as he first activates cells and leaves a trail of dots through the zone or to the point in the zone where he stopped or hid.

13.5.3.2 Trace Mode

In the trace mode, a bright dot is displayed in the picture on the alarm monitor at the center of each activated cell as in the normal mode. In addition, each illuminated dot emits a quick burst of flashes 8 seconds after it is activated. With manual reset, this results in a continuous moving trail of flashes at 8-second intervals along the path of intrusion. With automatic reset, a single burst of flashes occurs before each cell is automatically reset. These flashes can assist the operator in determining the size, direction, and location of an intrusion.

Larger monitoring sites require more cameras and mon-itors and a more comprehensive DVMD digital system. A high-speed microprocessor analyzes detected motion for size, position, and rate of movement to discriminate against undesired targets and to verify a valid intrusion

before the system signals an alarm. Verified intrusions ini-tiate audio and visual alarm signals. Video from alarmed cameras are connected to outputs for an alarm monitor, a recorder to monitor and record the track and position of intruders. Independent output relays provide control of external devices. A built-in sequential switcher provides normal system viewing of all cameras by separate video output.

For ease of operation, some systems have user-defined detection of active areas initiated using a light pen. Zones can be individually deactivated while observing the pic-ture to eliminate detection of areas where insignificant or acceptable motion could cause some false alarms. The systems have the ability to perform target discrimina-tion. Each camera module is programmable to optimize target discrimination based on a combination of antic-ipated characteristics, such as size, rate of movement, and indoor/outdoor scenes. In order to see the intrusion track and position display, zones where motion has been detected are highlighted on the video displays.

The system microprocessor analyzes the cell data and removes background clutter and identifies any changes in the cells as targets to be tracked. The target’s motion, speed, direction, and distance traveled are analyzed to see if they match the characteristics of a human intruder. When a human intruder is identified, on-screen graphics highlight his position and an alarm is signaled.

Special setup graphics define the camera zones to be monitored. Target discrimination is based on target size, contrast, speed, and direction. Target tracking is used to verify detection before declaring an alarm, resulting in a low false-alarm rate. The operator sets up sharply defined detection zones configurable for each camera, which may be tailored to reflect the optical differences between near and distant areas and act as distance compensation.

Digital video motion detectors are available in sizes suitable for small to large video surveillance systems (Figure 13-8). A family of products available is suitable for a single channel or four channels all provided with DSP electronics and microprocessors to analyze the entire video scene up to 30 fps for precision video detection of motion. At each update the system measures the pre-cise change in each pixel’s gray-scale level, i.e. the change in light intensity. These DVMD units are small in size, easy to install, and have simple pushbutton access for on-screen menu programming. They have access codes and password protection to protect against unwanted changes in programming by unauthorized personnel. The motion detection criteria include duration of motion and sensi-tivity. There are 99 levels of sensitivity permitting use in a variety of lighting situations. The 4, 9, and 16 channel units have built-in sequential switchers and provide alarm and video output from alarmed cameras. Alarm outputs can trigger TL VCR and DVR recorders, matrix switchers, quads, video printers, or video transmission devices.

Video Motion Detectors

367

One system has the ability to cascade up to 16 of the single channel units via a single host RS-232 serial port (Figure 13-8d). Figure 13-9 shows a block diagram of the multiple camera VMD system. One DVMD digitizes the scene by creating up to 16,000 individual zone locations per scene in up to 16 camera scenes. With this high resolving power, the system can detect an intruder occu-pying as little as 0.01% of the area. The DVMD system operates normally with a blank monitor. When a camera receives or detects motion, an audible alert is sounded and the disturbed scene appears on the monitor. The DVRs are activated for recording the intrusion scene or for reviewing the alarmed scene at a later time. When the DVMD displays the picture on the monitor, the guard sees the intruder in the scene even though he occupies only a small portion. The guard will also know where the intruder is, even if he is hidden from camera view, since the system displays the intruder’s path on the monitor. This display is accomplished by displaying bright flashes on the monitor at all locations the intruder has passed through. The guard now knows not only which scene was intruded upon but also the exact location of the intruder in that scene at that instant. He can therefore concentrate immediately on what decision to make and what action to take.

There is no industry standardization for the design and specifications of AVMD or DVMD systems. The fea-tures of some representative VMD systems and specific attributes are described in the following sections and Table 13-1.

13.5.3.3 DVMD Graphic Site Display Maps

An auxiliary display useful with VMD systems is an illumi-nated graphic display consisting of an overlay that is a plan view diagram of the entire monitored site. The map over-lay shows the location of each camera and alarm sensor, and flashes on the display when an intrusion occurs. To ensure that no intrusion is missed, particularly if there are simultaneous intrusions or motions in the scenes, video recorders are used. The recorder records the video scene, the intruder, his track through the scene, as well as a graphic alarm map if available. In the event of multiple video alarms in a single recording system, the recorder is set to record one alarm scene for a predetermined time interval and then switch to the next alarm scene. If a non-video sensor detects an alarm, the system acti-vates the appropriate camera(s) and the recorder. The displayed information enables the console operator to assess the situation rapidly and accurately and report any diversionary tactics. Present DVMD equipments are able to detect 20 times the number of intrusions as those detected by a guard looking at the video monitor without the benefit of the DVMD. This DVMD system is not easily mesmerized!

368 CCTV Surveillance

(B)
(A)

(C) (D)

FIGURE 13-8 Single and cascaded-single digital video motion detector (DVMD)

SECURITY OPERATOR

CONTROLS

CAMERA

1

VIDEO OUTPUT

CAMERA

MENU

VIDEO

MONITOR

VMD

VIDEO

PROGRAM

LOGIC

OUTPUT

RECORDER

SYSTEM

A/D

INTERFACE

INTERFACE

PRINTER

CAMERA

2

CAMERA

ALARM OUTPUT

VIDEO

VMD

AUDIO

LOGIC

A/D

VMD

DATA

VISUAL

SYSTEM

OUTPUT

KEYBOARD

CONTROLLER

DEVICE

INTERFACE

SECURITY

COMPUTER

N

LOGGING

PRINTER

ALPHA/

GRAPHICS

NON

NUMERIC

DISPLAY

VIDEO

OVERLAY

GENERATOR

ALARMS

ALARM

TIME/DATE

INTERFACE

GENERATOR

FIGURE 13-9 Multiple camera digital VMD block diagram

The DVMD analyzer detects the alarm condition by storing the scene in solid-state RAM. In one system, the storage process takes approximately 33 milliseconds and consist of sampling the picture scene (up to 16,384 dis-

crete locations) that are spaced throughout the scene. At each location the brightness is measured (one of 256 dif-ferent gray-scale levels). The address (pixel location in the scene and camera) is stored with the brightness number.

DISPLAYED VMD *
INFORMATION

ACTIVE AREA

MASKED AREA

MOTION ALARM

LOCATION OF ALARM:

SIZE OF MOTION AREA

(H × V) PIXELS)

MOVEMENT OF ALARMED † AREA-TRACKING

SETUP PARAMETERS

SENSITIVITY **
SIZE OF ACTIVE AREA (ZONES)

(NUMBER OF H × V PIXELS)

NUMBER OF ACTIVE ZONES

SHAPE OF ACTIVE ZONE(S)

DISABLED ZONES (ZONE MASK)

(SIZE, SHAPE, NUMBER)

PROBABILITY OF DETECTION ** ALARM LEVEL

CONTRAST

Video Motion Detectors

369

FEATURES

ON-SCREEN SETUP MENU

VIDEO LOSS DETECTION

NTSC/CCIR/PAL FORMATS

CONTROL P/T/Z

ALARM INPUTS

ALARM OUTPUTS

PASS THROUGH VIDEO

· ON-SCREEN DISPLAY: VARIES WIDELY DEPENDING ON SPECIFIC EQUIPMENT

** WHEN SET UP OPTIMALLY:

TYPICAL PPROBABILITY OF DETECTION—BETTER THAN 96%

TYPICAL NUISANCE ALERM RATE—LESS THAN 2%/DAY

TESTS BASED ON INDUSTRY STANDARDS

· AVAILABLE ON SOME MODELS

Table 13-1 Digital Video Motion Detector (DVMD) Features

This occurs for all zones in the scene. After the bright-ness and location information are stored, a comparison process is initiated that compares the present live picture from the camera (which the camera generates 30 times a second) to the stored picture. Whenever there is a bright-ness discrepancy in any zone, the address of that particular zone location is also stored with its brightness value. Zone locations where these differences are caused by electrical noise or ambient scene motion such as blowing leaves, trees, or flags are processed out and are not considered as alarms. All scene areas where detection is not desired are removed or masked out.

When a sufficient number of zones change, an alarm is processed. The comparison process occurs across the entire scene 30 times a second. The alarm condition is established by counting the number of locations with dif-ferent values; if a preset threshold count is reached (any number, but generally 1 in 8 counts), the system then alarms. The count is cleared each time a new storage process takes place. The memory is refreshed on a pre-set basis and ranges from 1/15th of a second to many seconds. Memory refresh prevents normal changes, such as scene lighting, moving clouds, or electronic drifts in the camera from being interpreted as alarm conditions. The camera viewing the intrusion scene is automatically switched to the monitor (any standard video monitor) and the scene displayed. The monitor is usually blank prior to an alarm, since there is no reason to display the scene if no activity is occurring. Table 13-2 summarizes the parameters of several commercially available digital VMD systems.

13.5.4 Features

VMD technology is not standardized, and therefore selecting the appropriate VMD approach requires under-standing the VMD features available and requirements of the application. Basic motion detection typically recog-nizes any type of motion in the camera FOV. A single output then activates automatic call up to the monitor screens for the surveillance personnel and initiates auto-matic VCR or DVR recording. With the advent of LANs, WANs and the Internet, the video call up is no longer limited to cabled CCTV systems, but can be transmitted over these communications channels, or even wireless. Advanced VMD products enhance the concepts of basic VMD through the use of elaborate algorithms that search out detailed movement patterns, and only activate a sys-tem response under very specific conditions. These activity criteria include:

· Intruder Identification: Identifying unauthorized humans in specified areas of the video FOV.

· Environmental Compensation: Recognizing and ignor-ing wind-blown debris, animals, background traffic, etc.

· Counting: Recognizing a quantity of a particular object or number of persons moving through an area.

· Direction: Ignoring objects moving in one direction, while alarming for objects moving in unauthorized directions (no identification).

· Item Recognition: Activating when specific user-selected items are removed from, placed in, or passed through the FOV.

370 CCTV Surveillance

VMD TYPE

CAMERAS

MONITORED

SINGLE CHANNEL

1

SIXTEEN CHANNEL

16

USER SETUP

*

TARGET **

SENSITIVITY

CONTROLS

PARAMETERS

SENSITIVITY

MINIMUM AGE †

OBJECT SIZE

MINIMUM MOVE

OBJECT DIRECTION

(#OF CELLS TO

OBJECT COLOR

CAUSE ALARM)

OBJECT MOTION

TARGET SIZE

MINIMUM VELOCITY

AND PATTERN

RESOLUTION ‡ (PIXEL LEVEL)

720 × 486

260,000

720 × 486

260,000

INPUT/OUTPUT

SIZE (inch)

SIGNALS

VIDEO

SMALL

ALARM INPUT/OUTPUT

1.5×3.5×5

DRY CONTACT

RS232, 422, 485

5 inch

CONTROL P/T/Z

RACK

OBJECT SPEED

MAXIMUM VELOCITY (PER CHANNEL)

MOUNT

· EITHER DONE VIA FRONT PANEL CONTROLS OR THROUGH SOFTWARE AND COMMUNICATION PORT

· TYPICAL PROBABILITY OF DETECTION–BETTER THAN 96% TYPICAL NUISANCE ALARM RATE–LESS THAN 2%/DAY STANDARD TESTS BASED ON INDUSTRY STANDARDS

· NUMBER OF FRAMES A TARGET MUST BE TRACKED BEFORE IT GENERATES AN ALARM. RANGES BETWEEN 1–300 FRAMES

· EACH ZONE IS COMPRISED OF A “BLOCK” OF PIXELS DEFINING THE ACTIVE OR MASKED ZONE

Table 13-2 Digital Video Motion Detector (DVMD) System Parameters

· Subject Tracking: Highlighting and following a specific person or item as it moves about the FOV or from the FOV of one camera to another.

· Multiple Subject Tracking: Highlighting and follow-ing multiple persons or items simultaneously as they move about the FOV or from the FOV of one camera to another.

13.6 GUIDELINES, PROS AND CONS

Some basic questions to be answered where VMD is required:

· Detection: Is there anything there?

· Classification: What is it—a car, person, bird, boat, van?

· Location: Where is it?

· Identification: Is it an unauthorized person?

· Is the person in the correct location at the site?

The security director and managers of a facility and the design professional who understand the VMD hardware options should begin a project by asking several important questions:

· What can move in the video image?

· What do we want to know about its movement?

The first objective is to identify what can move. This deter-mines the surveillance areas to be covered by the cameras and begins to define the VMD product required. The answer to what can move includes items of interest and any moving background items that may distract the sys-tem. The items of interest can be items that are typically in motion, and therefore either passed through the FOV or stopped in the FOV, these items require identification or must be followed by the surveillance cameras. Some of these moving targets include:

· Vehicles moving through entrances or a prescribed traf-fic route

· Routine entry and exit by unauthorized personnel

· Baggage left unattended

· Personal property that is carried by the public

· Suspect individuals

· Employee work methods or handling of assets.

Other items or activity that should be of concern include:

· Intruders or unauthorized personnel in an area or perimeter

· Leaks or mechanical failures

· Smoke, fire, or flame

· Violent or erratic behavior

· Counter-flow directional movement.

After the type of movement is understood, the next crite-ria affecting design and selection should be: What action should be taken when the motion of interest occurs? Does the alarm of interest require immediate response? If the incident requires immediate response, active surveillance personnel must receive the image and understand what they are seeing. They must also have instructions as to what action to take for each type of alarm. If the primary purpose of the video is for documentation or prosecution or litigation, changes in the FOV to accommodate the movement should be minimal, and more cameras should be implemented to confirm the events. In order to min-imize controversy and to allow acceptance in court, the graphic enhancement of the VMD, the storage methods for the video, and the signal compression methods must be closely scrutinized.

What should be the response and what action, if any, is warranted on the part of the officer based on the infor-mation presented? Video-based motion detection systems are providing many of the answers and solutions to this question.

13.7 SUMMARY

The primary function of the VMD is to allow the security force to make optimum decisions about an intrusion or unlawful activity in a minimum amount of time. Profes-sional intruders and thieves use devious and sophisticated techniques, making the guard’s response more complex. The intrusion scenario works to the advantage of crimi-nals because they can spend time planning it, as well as anticipating the guard’s action under duress.

The DVMD is a sensitive and valuable video security tool since it provides security personnel the visual information taken at the intrusion location when there is motion in the camera FOV. The intrusion scenario can be displayed on a monitor(s), recorded on a VCR or DVR, printed out on a hard-copy video printer, or transmitted to a remote site over a network.

The use of a DVMD significantly increases the security level and reduces the human error in any security system. The choice of the optimum VMD for a specific application requires that the security designer understands the equip-ment capabilities and limitations and match them to the problem. Of highest importance is whether the VMD can properly react to the changing lighting conditions in the video scene and generate meaningful alarm information and reject false alarms.

Video Motion Detectors

371

The present state of the art indicates that AVMDs can operate acceptably only in well-controlled indoor environ-ments, while DVMDs can operate in all indoor environ-ments and do well in most outdoor environments. Because of the variety of approaches and differences in DVMD equipment, characteristics of systems manufactured by leaders in the field must be considered on their own mer-its. Analyzing the systems described exposes the designer to some of the features available and permits asking the manufacturer sensible questions to determine suitability for the problem to be solved. Some helpful comments and hints follow:

· AVMDs or DVMDs are suitable for indoor applications.

· DVMDs should be used for all outdoor applications.

· The VMD should be able to switch video to a VCR or DVR and produce a hard-copy video printout.

· Once the VMD system is set up, most of the decision-making should be automatic.

· Following initial setup, alarm declaration should be automatic, using a menu-driven program.

An important axiom to remember is that the application should define the system rather than the system defining the application.

Chapter 14

Dome Cameras

CONTENTS

14.1

Overview

14.2

Speed-Dome Background

14.3

Fixed Dome

14.3.1

Technology

14.3.2

Housing

14.3.3

Hardware

14.4

Speed Dome

14.4.1

Technology

14.4.2

Housing

14.4.3

Hardware

14.5

Dome Mounting Hardware

14.5.1

Fixed Dome

14.5.2

Moveable Speed Dome

14.6

Cabling-Video Signal and Controls

14.7

Special Features

14.8

Special Applications

14.8.1

Outdoor Building Mounts

14.8.2

Pole Mounts

14.9

Summary

14.1 OVERVIEW

Fixed and Speed-Domes. The fixed dome camera has found widespread use in the video security industry. It has a monochrome or color camera and a fixed focal length lens. The camera is often mounted on a simple manually adjustable pan and tilt mount and the entire assembly mounted on a wall or ceiling.

The pan/tilt speed-dome has become one of the most popular scanning video camera surveillance system in the industry. The primary reason for their popularity is in the large amount of visual intelligence they can provide to the security operator in such a small physical package. The speed dome can be mounted almost anywhere: ceil-ing, wall, building exterior and pole. High resolution light

sensitive color cameras and compact zoom lenses with auto-focus mounted in ultra-fast pan/tilt module make them very effective in most environments including retail stores, casinos, commercial and government office build-ings, warehouses, airports, highways, etc.

One technique used to combine the conventional sepa-rate camera, lens, housing, and pan/tilt video surveillance assembly is to integrate them all in a plastic dome. The dome housing is more discrete than most other conven-tional housings. The dome camera consists of a round or hemispherical clear or tinted dome in which a camera, lens, and a manual or motorized pan/tilt mechanism are housed. The ceiling-mounted, below-the-ceiling, and wall-mounted hemispherical dome looks totally different from the rectangular housing and other shaped housings, and blends in well with many architectural décors. Since the hemispherical dome is circularly symmetrical, it can be in a fixed position and the camera pointed in any direction to view the scene. A pan/tilt module in the dome can rotate and tilt the camera and lens while inside the confines of the dome. This differs from cameras mounted inside rect-angular housings where the entire housing assembly and the camera move as one unit, and the pointing direction is known to the observer below. If the dome is tinted then the person down at floor level viewing the dome cannot see the camera and lens, and it is possible to point the camera in any direction without the observer knowing it is there, or seeing it move. This capability can act as an additional security deterrent because the observer does not know when he or she is under surveillance. Domes are less obtrusive and generally accepted in any environ-ment. Bullet cameras (commonly called bullet or lipstick cameras) are smaller and less noticeable, but they are visu-ally directional and the viewing and pointing direction is visible.

The moveable speed-dome camera contains an inverted pan/tilt mechanism suspended inside the dome with an

373

374 CCTV Surveillance

integral zoom lens and video camera module. The dome enclosure containing the camera/lens and pan/tilt mech-anism eliminates the precipitation, wind loading, dust, and dirt problem. The dome pan/tilt design is adaptable for use in outdoor applications on poles in parking lots, on building parapets, and under building eaves and passage-ways.

Indoor and outdoor, fixed and movable camera dome systems are available in many sizes ranging from 5 to 15 inches in diameter depending on the model. Fixed domes with miniature cameras and fixed lenses can be small and discreet and can have a manual pan/tilt adjusted during installation. Speed-dome systems use high-resolution color and monochrome CCD cameras with auto-focus and digital zoom, and zoom lenses.

The dome systems include camera pointing presets for pan, elevation, zoom lens focal length, and other param-eters. Another feature some dome systems have is privacy zone blanking that allows specific sections of the camera scene to be masked so that the operator cannot view scenes at pre-programmed camera pointing angles and zoom lens ranges. This prevents viewing the windows of private homes, hotels, or other buildings in the vicinity of the camera, as well as secured and classified areas. The zoom lenses and electronic zoom in the dome cameras can pro-vide powerful zoom capability with magnifications up to 200 times using electronic and optical magnification. The systems have sensitive CCD cameras that provide excellent color viewing during daytime operation and more sen-sitive monochrome viewing during nighttime operation. Dome cameras can be equipped with VMD and can send an alarm signal to the operator if there is movement in the image when it is viewing a fixed display.

14.2 SPEED-DOME BACKGROUND

There are essentially two types of camera systems that allow the operator to pan, tilt, and zoom the video image onto the monitor. The first type of system has been in use for many years, and uses a fixed camera and zoom lens mounted on a motorized pan/tilt mechanism. The electronics required for communications with the camera and platform motors and switches are installed in a sepa-rate enclosure. This type of pan/tilt platform is assembled from separate components and different manufacturers and has several shortcomings:

· The pan/tilt system is bulky and heavy.

· The camera pan/tilt pointing motion is slow—usually less than 10 /sec.

· The camera motion is usually restricted by the inter-connected cables, reducing the panning range below 360 .

· The cost for this type system is usually more than the newer speed-dome technology that uses an integrated camera, lens, and pan/tilt all in one dome assembly.

The new high speed dome systems employ newer more sophisticated technology having performance character-istics far superior to the older pan/tilt camera platform system. These speed-dome cameras are small in size: 5−7 inches in diameter and contain all the required con-trol and communication electronics located inside the unit. The dome module weighs far less than the older pan/tilt platforms so that they can be mounted almost any-where. The panning speed is typically 300−360 /sec and there are no interconnecting cables so that the cameras can be continuously panned without reversing direction. There are various manufacturers that can provide prod-ucts that have these basic functions.

Prior to the integrated dome with PTZ, pan/tilt plat-forms were assembled by ordering a housing, a pan/tilt mechanism, a camera, a lens, and wiring them up before installation. In the early 1980s, Sensormatic Inc. made a marketing decision to go into the video speed-dome market. With some of the initial concept coming from a company they had acquired, a large dome system using slip-rings to allow continuous 360 rotation and using a mirror—to reflect the incoming image onto the lens and produce a lower profile dome system—was built. The sys-tem also integrated the receiver driver portion of the PTZ control electronics into the dome assembly. To improve accuracy of pan and tilt and increase the speed sub-stantially, stepper motors replaced the AC motors. This also made possible the incorporation of dome pan/tilt pointing presets into the system for defining targets, pat-terns and boundaries. The entire dome was assembled and tested in 1985, and represented one of the first fully integrated “speed” domes. The first system had a clear viewing bubble 22 inches in diameter and was 8 inches deep. The system used a monochrome vidicon tube cam-era and weighed approximately 40 lbs. In 1988 a second-generation speed-dome using a color CCD camera imager was produced. The bubble was reduced to 12 inches in diameter and 5 inches deep and weighed 26 lbs. To get the smallest size for the system the CCD sensor and lens were located remotely from the camera body using a high-flex cable. This produced a very short camera-lens assembly. In 1992 the speed-dome received a complete mechani-cal redesign and used a close-loop DC servo electronic pan/tilt design providing the ability to point to any target in less than one second. The first application for these sys-tems was in casinos and interfaced with American Dynam-ics matrix switchers.

A second pioneering company in the speed-dome field was Diamond Electronics, producing a high-velocity rate-proportional digital tracking system. It had a slip-ring design to permit continuous 360 rotation at speeds up to 80 /sec and tilt speeds up to 25 /sec. Dynamic brak-ing featured immediate precise stops with ±0 5 accu-racy when de-accelerating from any speed. The system had gold and chrome tinted dome capsule enclosures providing one way mirror capsules providing for discrete

surveillance. These dome capsules were optically corrected for high-performance monochrome and color camera sys-tems. The drive electronics and camera electronics were all contained within the dome package.

The current and latest generation speed-domes are available in sizes from 4.5 to 10 inches in diameter and have variable high-speed pan/tilt stepper or servo motor drives with continuous 360 rotation obtained with metal or optical slip-rings. Camera-pointing features include Pre-sets, Patterns, and Boundaries. The cameras include high-resolution daytime and nighttime capability using color for daytime and switching to monochrome for higher sensitiv-ity during nighttime operation. Camera features include VMD and alarm activation on motion. These camera-lens modules are equipped with motorized zoom lenses with optical and digital zoom, auto-focus, and iris control. Zoom lenses have 20:1 zoom ratios to obtain telephoto and wide-angle viewing. The camera pan/tilt module is mounted within a rugged, clear, or smoked hemispheri-cal optical grade acrylic plastic dome designed for quick installation, mounting, and servicing in a ceiling, a wall, outdoor on a building parapet, a parking lot pole, or on a highway.

14.3 FIXED DOME

The fixed dome camera assembly has become a very attractive enclosure for providing surveillance in almost any environment. The nature of the round dome with a smoked or tinted dome makes it unobtrusive and does not allow the observer to determine in which direction the camera is viewing. There are many manufacturers pro-ducing fixed video domes with fixed camera or manu-ally adjustable pan/tilt mounts. The cameras provided are monochrome or color and the lenses have FOVs from 90 wide-angle to 30 narrow-angle providing an inexpensive, attractive integrated camera for most indoor applications. Some models have variable focal length (vari-focal) lenses to make it easier to obtain just the right camera FOV. For outdoor applications, larger domes with larger and longer focal length lenses are available to provide sufficient mag-nification for the longer distances. These outdoor domes are available with thermostatically controlled heaters and fans and are sealed against moisture and the environment. All these fixed dome cameras are available with infrared LED to provide operation in nighttime at distance up to 20 feet without any auxiliary lighting.

14.3.1 Technology

The fixed dome cameras use monochrome and color CCD or CMOS cameras with lenses to view narrow-angle, medium- to wide-angle FOVs under most lighting condi-tions. Typical sensitivities are 1–2 lux for color cameras and

Dome Cameras

375

0.1 lux for monochrome. Resolution is typically 480 TV lines for color and 570 TV lines for monochrome. When there is not enough or no lighting an infrared LED cam-era is used. Many dome manufacturers mount the camera so that it can be manually adjusted in the horizontal and vertical (pan/tilt) directions. These fixed domes are small and lightweight and are easily mounted onto a drop ceil-ing, hard ceiling, or a wall. Dome cameras are available with standard analog signal outputs for use with coaxial cable, unshielded twisted pair (UTP) or to interface with other transmission means. There are IP network dome cameras that can be connected directly to a LAN, WAN, or the Internet.

14.3.2 Housing

Most indoor fixed dome housings are manufactured using ABS or polycarbonate plastic. The lower dome bubble through which the camera lens views is manufactured from optically clear acrylic plastic. Most systems are provided with a clear, tinted, or smoked plastic bubble. Special variations include bronze, chrome, and gold. The clear bubble essentially transmits all of the light and is used when maximum light throughput is required. The smoked dome loses about 30% of the light (70% transmission), the bronze approximately 50% (50% transmission) of the light, and the gold approximately 75% of the light (25% transmission).

Outdoor housings are available with UV-protected ABS or vinyl, or painted aluminum or steel. For harsh or extreme environments or where corrosive atmospheres or severe vandalism is present, dome housing materials are fabricated from polycarbonate plastic, machined or cast aluminum, or stainless steel.

14.3.3 Hardware

There are many manufacturers of fixed dome camera sys-tems. Figure 14-1 shows examples of indoor and outdoor fixed dome cameras. The size of these domes varies from 4 to 6 inches in diameter and weigh from 1 to 2 lbs. They are available with clear or smoked viewing domes.

14.4 SPEED DOME

The majority of conventional camera/lens pan/tilt plat-forms in housings consist of components obtained from several different manufacturers all assembled by the sys-tems integrator and made to operate as a complete system. This is a practical solution for installations in which the parameters and characteristics of the fixed or movable dome might be unacceptable. The dome camera inte-grates the camera/lens, pan/tilt, housing, and mounting

376 CCTV Surveillance

FIGURE 14-1 Fixed dome camera

systems

(A) FIXED COLOR DAY/NIGHT (B) FIXED COLOR INTERNET (IP)
ANALOG CAMERA MPEG/JPEG CAMERA

system in a single module from a single manufacturer. The integral design results in a smaller, lighter weight module having a high scanning speed and wide angular coverage.

Speed-dome systems can scan at a rate of 300 /sec and are capable of panning 360 continuously using slip rings. With 360 continuous horizontal scanning the lens/camera module does not have to come back 360 in order to follow a moving target. All the components can be housed in a 5−7-inch diameter ceiling-mounted dome. Through advanced engineering and compact packaging, these fast scan rates were obtained: the moving parts are small in size, and have low masses and moments of inertia. The obvious advantage of a fast system is that if an incident occurs anywhere within the dynamic FOV of the pan/tilt system, the camera/lens can be pointed in any direction in the shortest possible time while the lens zooms and focuses on the target. Microprocessor-based dome systems with camera-pointing preset capabilities can take advan-tage of these fast pan/tilt designs.

14.4.1 Technology

The speed-dome assembly contains a high-speed pan/tilt assembly, high resolution day (color) or night (monochrome) CCD camera with a compact 20:1 zoom ratio lens with continuous full-time auto-focus function. One system has a wide dynamic range feature that can pro-vide detailed images when the camera is viewing images that have bright light and low light level image areas.

Camera, Lens. Most speed-dome systems use high sen-sitivity color cameras that can be: (1) operated in color,

· operated in monochrome, or (3) switchable from color to monochrome automatically. The CCD cameras have an image format of 1/4 inch, and along with a compact zoom lens, provide a small compact design resulting in high pan-tilt speeds. Overall camera resolution is typically 480 TV lines for color and 570 TV lines for monochrome.

Values of 1 lux sensitivity for color and 0.06 lux or less for monochrome are typical. One system using a patented signal level compression technique can provide images that have over 60 times the dynamic range compared to other cameras. Cameras are also provided with automatic brightness compensation (ABC) so that the camera can view scenes containing both bright and dark areas. This overcomes the problem that if a camera is located in a poorly illuminated room and pointed at a window with a brightly illuminated scene outside, the camera will either set its iris level to optimize the inside or outside scene. This results in one part of display being normal while the other part is either too light or too dark. This also occurs in the evening when viewing oncoming traffic with the headlights turned on. The ABC enables the camera to see both the light and dark areas of the display with reduced flair from the oncoming headlights.

The zoom lenses generally have a 20 to 1 optical zoom (magnification) range that is extended by electronic digi-tal zoom by another factor of 10 providing an overall 200 to 1 zoom range. Sensitivity of the color cameras are down to 1 lux for color and to .05 lux for monochrome. Switchover from color to monochrome is automatic when the light level falls below a predetermined level. To capture image detail in both light and dark regions, Panasonic Inc. uses the Super Dynamic SDII technology which records the scene at two different exposures and then electronically integrates both of them into a single image to preserve the detail throughout the bright and dim areas. This added to additional precise color reproduction creates a dynamic range that is about 64 times greater than that of conven-tional cameras.

Pan/Tilt Mechanism. The speed dome panning mecha-nism provides 360 of continuous horizontal panning rota-tion. To obtain the 360 rotation slip-rings are used. Some systems use a light transmitter and receiver to transmit the signal information rather than a metal slip ring assem-bly. The tilt mechanism provides for at least a 90 vertical range of travel. In most cases the camera assembly can tilt

up above the horizontal a few degrees and down −95 to provide a tilt range beyond looking straight down to look-ing slightly above the horizon. Precise manual panning and tilting is achieved through a combination of a variable speed control in the form of different speed ranges, with an automatic adjustment of the speed range depending on the zoom position of the lens. For wide-angle zoom-ing the speed is increased, whereas for high magnification (telephoto zooming) the panning and tilting speeds are decreased. Depending on the manufacturer, the panning and tilting are done using DC servo motors or stepper motors. To provide high torque and precise pointing abil-ity, the DC servo design uses pulse-width modulation and speed feedback to control the acceleration, speed, and de-acceleration of the motors, ensuring a smooth, precise, accurate, and fluid movement. Most manufactures design the drive systems so that there are no belts or pulleys insuring long-term reliable operation.

Dome Cameras

377

An example in which panning speed is important fol-lows: A person walks past a dome pan/tilt unit 15 feet away the dome (Figure 14-2). If the person is walking at a normal rate of about 4 feet per second and the dome is panning at a rate of 1 foot per second (12 /sec), the monitor scene at 15 feet is moving at a rate of 3 feet per second. The subject is quickly lost because the pan/tilt cannot pan fast enough to follow the subject. With a high-speed, 60 /sec panning system, a target at 15 feet from the camera produces a picture going by at a rate of 5 feet per second (1 foot per second faster than the target), and the subject is not lost. In this example, the panning speed would be reduced to 4 feet per second to keep the target in the center of the picture.

Slip-Rings. Most standard pan/tilt platforms use a mech-anical stop at each end of the horizontal and vertical pan-ning ranges to prevent the wires connected to the moving

· LIMITED 355°

CONVENTIONAL PANNING

5 CONTINUOUS 360°

HIGH SPEED PANNING

360° PANNING MOTION

60°/sec MAX

15

PERSON WALKING AT

4 ft /sec

P/T PLATFORM MUST STOP

AND REVERSE DIRECTION

TO REACQUIRE MOVING TARGET

EVEN AT 24°/sec
IT REQUIRES 15 sec TO
ROTATE 360° AND

REACQUIRE TARGET

MAX PANNING SPEED 60°/sec

PRODUCES 5 ft /sec @ 15 ft

FROM THE CAMERA

15 ft

LENS

FOV
T = 4 sec

PERSON WALKING AT

4 ft/sec

LENS

FOV

T = 0

FIGURE 14-2 Target speed vs. panning speed

378 CCTV Surveillance

camera/lens assembly from getting twisted (the wire ends are terminated in the stationary wall mount). This means that the camera cannot scan more than 355 horizontally before it must stop and then pan in the opposite direc-tion. Even at a 24 /sec pan speed, nearly 15 seconds is required to acquire a subject or target that is moving past the end of the panning range. During most of the 15 sec-onds the target is out of sight of the camera and probably lost. The speed dome camera does not have this limita-tion as it continues to follow the target. This is one of the salient reasons why speed dome systems are such effective surveillance cameras and have replaced many pan/tilt platforms.

In the panning system using slip-rings, the camera/lens combination rotates continuously and beyond 360 with-out any concern for twisted wires, since the electrical sig-nals and power pass through the stationary slip-rings. No matter where the target moves in the lens FOV, the pan-ning motion can continue: the subject never leaves the FOV. There are no restricting mechanical stops to limit the pan/tilt unit’s rotation.

Most dome manufacturers use gold plated metal slip rings to transfer the video control and power signals from the camera to the dome base and on to the communica-tion channel. Others use optical slip rings for the video. The all-optical connection between the moving camera and the base can provide a higher quality image with less video noise than the metal gold contacts. Transmission of the video signal by a light that requires no physical con-tacts makes for a better “slip ring.” This eliminates the possibility of image noise and enhances the reliability of the dome unit.

14.4.2 Housing

Most indoor speed-dome housings are manufactured from ABS or polycarbonate plastic. The lower dome bubble through which the camera lens views is manufactured from optically clear acrylic plastic. Most systems are pro-vided with a clear or smoked plastic bubble. Other tints available include bronze, chrome, and gold. The clear bubble essentially transmits all of the light and is used for maximum light throughput. The smoked dome loses about 30% of the light, the bronze approximately 50% of the light, and the chrome (aluminum) and gold approx-imately 75% of the light. Only the clear and smoked versions are generally used for outdoor applications.

If the camera/lens pointing axis is not perpendicular to the dome surface (Figure 14-3) and looks at an oblique angle the images may appear elongated vertically or hor-izontally. If the dome and camera are in a fixed position with respect to one another, the distortion is generally less noticeable than if the lens is panning or tilting while the dome remains still. Figure 14-4 shows widely used dome housing configurations.

For outdoor applications, the domes are equipped with thermostatically controlled heaters, blowers, and protec-tive sun shrouds. Standard housing colors include gray, white, or black baked on enamel. The lower domes through which the camera views are clear or gray (smoked).

14.4.3 Hardware

There are many manufacturers producing high speed dome camera systems. Table 14-1 shows some of the

CENTER

CAMERA

OF DOME

OFF CENTER

360°

VERTICAL TILT

–90°

HORIZONTAL PANNING

FIGURE 14-3 Camera viewing through dome

Dome Cameras

379

FIGURE 14-4 Representative

speed-dome systems

11 480 TVL COLOR CAMERA

18× OPTICAL MAGNIFICATION MPEG/JPEG INTERNET (IP)

10 510 TVL COLOR CAMERA

22× OPTICAL MAGNIFICATION

ANALOG OUTPUT

17 COMPACT SELF CONTAINED CCTV PLATFORM

18 INTEGRATED, ENVIRONMENTALLY CLOSED DOME

19 HIGH SPEED PANNING —200°/sec

20 HIGH POINTING ACCURACY: ±0.1°

21 360° CONTINUOUS PANNING

22 COMPACT, UNITIZED CAMERA/LENS/PAN/TILT MODULE

23 HIGH ZOOMING (MAGNIFICATION) RANGE:

OPTICAL: 10–20×

DIGITAL: 10–20×

OVERALL: 10–200×

24 AUTO-REVERSE FOR DOWNWARD VIEWING

25 PRESETS: PAN, TILT, ZOOM, TOUR

26 MENU-DRIVEN-REMOTE SETUP

Table 14-1 Key Features of Speed-Dome Systems

features of speed-dome systems available. Section 14.7 describes many extra features not described in Table 14-1.
The two high-end designs by Pelco and Panasonic rep-resent the most complex and full-featured systems. With technology advancing regularly, these systems will contin-ually be updated and supersede the capabilities of those listed in the table. Most of these systems have many fea-tures in common but with different specifications (see Section 14.7 for additional features). Table 14-1 briefly outlines the key features of speed domes. Most contain a color camera that is switched electronically, or mechani-cally moves an optical filter in or out of the image light path to the camera sensor. They contain a high-speed pan/tilt servo or step motor drive system, and a clear, smoked, or other tinted viewing material. The high-quality dome material is of high-quality acrylic and is optically clear with no distortion in any portion of the dome that is viewed through by the camera/lens. These domes are available for indoor wall mounting, ceiling mounting either recessed or as a pendant on a building, or pole

mounted. They are available for outdoor applications for parapet building mounting, on fixed poles in parking lots or highways. The panning speed for most speed domes varies from 0.1 to 360 /sec continuous rotation. The verti-cal tilt ranges from +2 above the horizon to −92 below the horizon. These systems have manual override for speed control that ranges from 0.1 to 80 /sec in panning to 0.1 to 40 /sec in tilting. In the automatic preset mode, the panning speed can be up to 360 /sec and the tilt speed up to 200 /sec. Most speed domes have capability of pro-gramming presets including the ability to select auto-focus modes, iris level, and light compensation. Some systems have the ability to copy a preset command from one cam-era to another. Programming can be via keyboard through the dome system on-screen menu. Preset accuracy can be as low as ±0 1 . These systems are provided with limit stops that are programmable and used when the operator uses manual panning. Most are provided with an opaque mechanical/optical liner that rotates with the dome to ensure that the camera and pan/tilt assemblies are not visible to the observer. The domes are available with alarm inputs and outputs. Programmable patterns can be user-defined including pan, tilt, and zoom for the preset point-ing directions. The security manager can block out spe-cific areas and specific viewing directions to eliminate viewing secured areas and areas requiring privacy. The domes are almost all available with a menu-driven setup and operational modes. The menus can be displayed in different languages for initial installation and operational use. Many have an image flipping feature that inverts the dome image 180 at the bottom of the tilt travel, so that the image is always right-side-up when the camera view-ing angle passes through the vertical downward rotation. Depending on the system, communication to and con-trol from the monitoring console is performed through multi-conductor cable, coaxial cable, UTP, fiber-optic, or

380 CCTV Surveillance

third-party control systems. Video motion detection is available on most systems when in the preset mode of operation, with alarm outputs activated. Most indoor and outdoor systems are fabricated using painted aluminum construction with outdoor systems available with stainless steel construction; either non-pressurized or pressurized models are available depending on the application.

Figure 14-4 illustrates some of the many standard types of speed-dome systems available. Since most of these domes are mounted at ceiling level, on a parapet atop a building, or on the top of a pole, they must be designed for easy installation and maintenance. Each has a unique quick-disconnect mechanical install and removal inter-face for mounting the dome section to the permanently mounted base section.

Pelco. Figure 14-4a shows a speed dome having a variable panning speed from 360 /sec continuous down to 0 1 /sec. The manual control range is from 0.1 to 80 /sec, and pan at 150 /sec in what is called turbo mode. The tilt speed ranges from 0.1 to 40 /sec. When in the automatic preset mode, the panning speed is up to 360 /sec and the tilting speed is up to 200 /sec. The vertical unobstructed tilt is from +2 above the horizon to −92 .

Panasonic. Figure 14-4b shows a speed dome with a color CCD camera having a 22 times zoom, auto-focus lens, and rotating chassis in a 4.3-inch diameter housing suitable for most indoor locations. It has an additional 10 times electronic digital zoom for a total zoom range of 220. The color camera operates at light levels of 1 lux and produces monochrome images at 0.06 lux. It has full 360 horizontal rotation and 90 vertical panning, and has a speed of 300 /sec. It incorporates digital motion detection for advanced alarm applications. The camera has 510 TV line horizontal resolution.

14.5 DOME MOUNTING HARDWARE

Many manufacturers produce attractive dome housings and mounting configurations for indoor and outdoor fixed and pan/tilt dome systems (Figure 14-5). For indoor applications, the domes are securely attached to a wall or ceiling mounting bracket. The electrical cables connected from the camera and the pan/tilt mechanism are directed into the wall or ceiling.

14.5.1 Fixed Dome

The fixed dome module consists of the camera, lens, and housing with dome and is installed on the surface of a wall, ceiling, building exterior, and pole with appropriate mounting hardware.

14.5.2 Moveable Speed Dome

The speed dome structure consists of two basic parts:

(2) the rear box which is installed or mounted on the mounting surface (wall, ceiling, and pole) and (2) the dome with the camera pan/tilt mechanism. Most manu-facturers use a quick, positive, mechanical, and electrical disconnect between the rear box and the camera/dome assembly that does not require the use of any tools. This is particularly important in retail stores, warehouses, parking lots, and highway applications since the dome is usually mounted at elevations requiring ladders or other means to reach the dome. This installation and maintenance issue has been addressed by several companies that now pro-duce dome systems that can be installed and maintained at ground level (Section 14.8.2). The domes for these pole-mounted systems are raised and lowered mechani-cally. The dome is brought down to ground level during installation or servicing and they are raised for operation at the elevated level at the top of the pole. For these video domes the pole is part of the dome system.

In harsh outdoor environments or for chemical protec-tion, type 316 stainless steel enclosures are available with a height of 11 inches including mounting and dome, and a 10 inch diameter. These enclosures require no painting and withstand all outdoor environmental conditions as well as having higher impact ratings that are each important when the systems are located in areas of vandalism or other attacks. Where required, pressurized stainless steel pen-dants are available with an overall height of 12 inches and an 11 inches diameter. These domes have Schrader type fill and pressure relief valves and operate at 5 lbs/square inch gage (psig) pressure typical and 7 psig pressure relief. These systems usually incorporate internal sensors for pressure, humidity, condensation, and temperature, and are usually equipped with heaters or blowers where the environment requires. These systems are equipped with internal sensors reporting with on-screen displays of sensor indications and sensor out-of-range reporting.

14.6 CABLING-VIDEO SIGNAL AND CONTROLS

The speed-domes communicate to the console and net-work via built-in multi-protocol receiver/driver assem-blies for use with matrix switching systems and other equipments. The types of protocols supported by many manufacturers include: (1) AD Manchester control code using a single 18 AWG shielded twisted pair (STP) to sup-port several daisy chained domes at a maximum of about 5000 feet, (2) 22 AWG UTP to support up to 32 daisy chained domes to a maximum of 3200 feet, (3) AD-UTC and RG-59U video cable to control a dome to a maximum of 1600 feet. These receiver drivers located in the dome provide all the voltage necessary for camera controls, pan and tilt functions, and all motorized lens functions. Most

Dome Cameras

381

(A) FLUSH CEILING MOUNT (B) PENDANT CEILING MOUNT

(C) PENDANT WALL MOUNT (D) FLUSH WALL/CEILING MOUNT

FIGURE 14-5 Indoor video dome mounting configurations

dome interfaces support selected third-party protocols for integration into other systems. These can take the form of fiber-optic communications or other types. The dome includes standard support for UTP dome connections that allows the use of CAT cabling for transmission of video or video up the coax dome control signals up to 1000 feet. Communication protocols provided by many manufactures include RS-422, RS-232, and RS-485.

There are several techniques for the console controller to communicate with and control the remote moveable speed-dome camera:

· Direct Wire—video coax with multi-conductor for con-trols

· UTP—video with multiplexed controls

· Single Coaxial Cable—multiplexed video and controls on coaxial

· Wireless—video and controls transmitted via RF or microwave.

Direct Wire. The simplest control of the PTZ lens mech-anism is via direct wire, using one wire for each control function and a separate video coaxial cable. This straight-forward technique is in widespread use for many small or short-run (under a thousand feet) installations. This technique requires no additional driver electronics for transmitting the control signals and no additional receiver electronics at the camera unit. The controller consists of switches that control all functions set manually by the oper-ator or memorized by the system for automatic operation. Wire size must be large enough to minimize voltage drop to the motors and electronics.

Unshielded Twisted Pair (UTP). For longer distances or when there are many different camera sites, a significant reduction in the number of conductors and wire runs is accomplished by multiplexing (time-sharing) the control signals at the control console onto two UTP wires, sending them to the camera site, and then de-multiplexing them or

382 CCTV Surveillance

separating them again to provide the signals necessary to drive the PTZ unit. Since the two wires need to carry only communications information and not current to drive the motors, any long-distance two-wire communication system suffices. Two popular transmission codes (protocols) are the EIA RS-422 and RS-485. The video signal is transmitted on a separate coaxial cable or UTP.

Single Coaxial Cable. Several companies manufacture systems that multiplex or time-share the control signals in video signal on the same video coaxial cable, thereby allowing video to be transmitted from the camera to the monitor console site, and camera control signals to be transmitted from the security console to the camera site, all on one coaxial cable.

For direct wiring, this is an efficient solution since only a single coaxial cable is required. The system requires a simple multiplexer that combines the video and control signals at the camera and the monitor ends. An advantage of multiplexing the control signals onto the video signal is that additional transmission or control signals can be added to the system without adding new cable. These addi-tional functions can include lens controls, alarm functions, or tamper switches.

Wireless. Control signals can be transmitted from the console to the camera location via wireless remote con-trol communication. The control signals are multiplexed onto a single channel and transmitted on RF, microwave, or light-wave (visible or infrared) communication links. In extreme security environments (such as military or nuclear sites), wireless transmission of video, command, and con-trol signals is used as a backup to a hard-wired (copper or fiber-optic) system.

Fiber-optic. The fixed and speed-dome systems have compatibility with fiber-optic transmitters used for long-distance cabling runs. Fiber-optic transmission is an alter-native to copper wire, and many manufacturers have equipment that transmits the control signals, alarms, and video signal on a single fiber-optic channel. As mentioned in Chapter 6, the fiber-optic advantages include noise immunity, long transmission distance, absence of ground loops, high security (difficult to tap), and reliable opera-tion from different building sites in harsh environments.

Third-party Communicators. The fixed and speed-dome systems have compatibility with and capability to be con-nected into optional boards that convert the control signals into a suitable form for the selected third-party controllers.

Digital Network. Fixed and speed-dome camera systems are now available that can be connected directly into ana-log or digital networks. When the camera is connected to a LAN, WAN, or Ethernet network, the operator can view and operate the system and monitor the images locally or remotely using a PC.

Wiring Access Panel. The installation of the dome base is normally accomplished prior to the purchase or instal-lation of the dome housing itself. The dome base should have an easy access door that allows complete access to the installation wiring, and when closed it should provide complete separation of this wiring from the dome drive.

14.7 SPECIAL FEATURES

Camera Sensitivity. Most dome systems have dual-mode cameras that operate in color mode during daytime and monochrome mode during nighttime. In addition, some cameras have the feature to provide temporary image enhancement under low light level conditions via manual override. This override reduces the shutter speed from the normal 30 fps to 2 fps resulting in a 15 times increase in camera sensitivity.

Memory. Non-volatile memory storage and location-specific dome settings such as presets and patterns are built-in for the camera. If a new dome drive is installed in the system, all the settings are downloaded automatically into the new dome drive.

Motion Detection. Domes support VMD within a preset. The motion detection trigger action includes activating a preset command, activating a pattern, and sending a dome output to the console.

Presets and Patterns. Most domes support camera pre-sets programmed into the dome module so that the dome can point (pan/tilt) to a preset direction. Models with as many as 96 presets and 60 patterns of presets are pro-grammable. Domes are also designed to support a Home Position that automatically returns the dome to a Preset, Pattern, or Preset Sequence after a specified period of inactivity anywhere between 1 minute and 1 hour. Also provided is a freeze frame function that maintains a static image on-screen during dome movement and lens adjust-ment when presets and patterns are called. This freeze frame function helps preserve hard drive space when a VCR or DVR is used.

The speed-dome parameters that can be preset include:

· auto-focus mode, (2) iris level, (3) back-light compen-sation, (4) the ability to command to copy the camera settings from one preset to another to reduce setup time, and (5) to preset programming the control keyboard or the dome system on-screen menu. The preset accuracy can be as high as ±0 1 .

Proportional Pan/Tilt Speed. The system panning and tilting speed can be increased or decreased depending on the instantaneous zoom focal length. To optimize the viewing of the image on the monitor for different zoom positions, when the zoom lens is in wide-angle position the speed is increased, and when it is in the telephoto (high magnification) position it is decreased, and proportionally optimized in between.

Digital Flip. The speed-dome should have a provision for quick image reversal that automatically pans the camera
· when the bottom −90 tilt limit is reached to allow for continuous tracking of a target passing directly beneath the dome. This is important when following a person who is passing directly under the camera from one side to the other.

The digital flip feature allows for more convenient mon-itoring when viewing objects that pass directly below the camera. As the camera pans in the vertical direction to follow the object, DSP automatically flips the image to the bottom as the object passes beneath the camera so that the image remains right-side-up for easier viewing. In addition, the system contains an image-hold feature that prevents blurring when the camera moves and does the 180 flip. It maintains the image prior to flip after the 180 flip.

Privacy Zone-Window Blanking. Some domes support privacy zones to prevent users from viewing sensitive or secured areas. So as not to interfere with normal surveil-lance operations, these on-screen shields must block out only the area that has been defined as sensitive. The pri-vacy cell should not cause the screen to blank out.

These privacy windows are available in: (1) four-sided user-defined shapes, (2) opaque gray or translucent smear,
· blank all video above a user-defined tilt angle, and

· blank all video below a user-defined tilt angle.

Zoom-Distance Compensation. Whether the dome cam-era is in the privacy zone or the lens is zooming from wide-angle to telephoto the system should compensate for a specific focal length in use at the time. For any specific focal length, the zoom lens should adjust the alarm or pri-vacy zone window to compensate for the changing FOV. This is called zoom-distance compensation.

Monitor Display, Menu. The speed-dome systems sup-port on-screen programming of the dome parameters including image flip, direction indicators and azimuth, maximum zoom stop, camera line lock or internal crys-tal synchronization, AGC, white balance, VMD selection, alarm actions and default states, and home position. They also display on-screen programming of: dome names, area names, preset names, pattern names, and alarm names. Most systems provide most of these attributes in English, French, Italian, German, and Spanish, as well as in other languages. The on-screen text characters are available as user-selectable in solid or translucent white, with or with-out a black outline.

Alarm Inputs. The dome assemblies have single or multi-ple alarm inputs as an option and are field programmable to receive normally open or normally close contacts. If the system is operating on an RS-422 network, the domes are capable of receiving the alarm and transmitting it back to the switching system, and/or reacting to the alarm event independent of the switching system. If a Manchester net-work is used, the dome is capable of processing the alarm

Dome Cameras

383

internally in the dome and automatically activating a Pre-set, Pattern, or Preset Sequence.

Twist Lock Release. Maintenance is an important factor to consider in ceiling or pole-mounted dome camera sys-tems. To simplify installing and servicing these domes, most systems contain a quick disconnect or twist lock release at the base of the dome. The standard base of the dome is hard mounted to the wall, ceiling, or pole mount and contains a receptacle for direct wiring to the dome assembly. All wiring is done before lifting the camera pan/tilt assembly onto place. The base assembly includes a tamper switch so that if the dome cover is removed, an alarm is sounded. The quick disconnect base allows wiring to be done once directly in place and then installing or servicing the dome assembly without disturbing any of the wires or connections. Normally each base includes diagnostic LEDs to indicate power and proper commu-nications to and from the console or matrix switcher. Some designs require a simple tool to remove the dome assembly; however, others require no tools and are simply installed or removed using a twist lock release. It is impor-tant that the dome and the base are available separately so that the installation of the base can be accomplished by the installer prior to the purchase of the dome hous-ing/camera assembly.

14.8 SPECIAL APPLICATIONS

The use of fixed and speed domes in elevated locations in buildings, on exterior walls of buildings, and outdoors, in general, has resulted in the design of many different configurations for mounting these domes.

14.8.1 Outdoor Building Mounts

Figure 14-6 illustrates outdoor speed-domes mounted on a building roof edge and capable of scanning 270 and

· horizontally to view parking garages and lots. With such a large angular FOV to cover (an entire parking lot), this solution should be used where only sporadic activity is monitored, since panning with a standard unit from one end of a building would not keep most of the parking lot under surveillance. Adding additional speed domes would increase coverage.

14.8.2 Pole Mounts

Figure 14-7 shows dome camera pan/tilt assemblies mounted on poles and pedestals to provide wide-angle video surveillance at entry and exit roadways, parking lots, streets, etc. Mounting the camera away from the building on a pole provides good viewing of the entire

384 CCTV Surveillance

FIGURE 14-6 Standard outdoor speed-dome mounting configurations

(A) STANDARD (B) BUILDING MOUNT

(A) PEDESTAL/WALL MOUNT DOME (B) CORNER WALL BRACKET (C) POLE/WALL MOUNT

FIGURE 14-7 Outdoor dome and mounts for buildings, roadways and parking lots

building entry area with a single camera. The presence of the camera system serves as a deterrent to crime while it captures the necessary visual information for possible apprehension and prosecution. The same scan-ning limitations as described in the previous system apply.

There is one disadvantage of the dome pole camera or any camera mounted on a pole: the difficulty of performing maintenance on it. Several companies have pursued designs that permit easier maintenance. The widespread use of the speed-dome in parking lots, on walkways, and on streets,

highways, etc. has motivated manufacturers to design inge-nious means to raise and lower the entire dome assembly from ground-level (Figure 14-8). The video dome in Figure 14-8a, b is raised and lowered using an electric drill.

14.9 SUMMARY

There are many varieties of camera housings and inte-grated camera systems for video surveillance applica-tions. The configuration that has become most popular

Dome Cameras

385

(C) HEAVY DUTY

(D) HEAVY DUTY

(A) RAISING AND LOWERING (B) NORMAL POSITION

FIGURE 14-8 Pole-mounted dome assemblies maintained from ground-level

is the dome housing that is available in a fixed or speed-dome configuration. These dome camera systems are suitable for indoor and outdoor applications avail-able with monochrome cameras, or color cameras that can automatically switch from color for daylight use to monochrome for extended low light level sensitivity and

produce optimum performance at most all light levels. The speed dome provides a very powerful video surveil-lance tool for gathering maximum visual intelligence and is in widespread use in retail establishments, casinos, ware-houses, outdoor parking lots, pathways, building exteriors, and streets and highways.

Chapter 15

Integrated Cameras, Camera Housings, and Accessories

CONTENTS

15.1 Overview

15.2 Indoor Housings

15.2.1 Functional Requirements

15.2.2 Indoor Types

15.3 Outdoor Housings

15.3.1 Functional Requirements

15.3.2 Outdoor Design Materials

15.3.3 Outdoor Types

15.4 Integrated Cameras

15.4.1 Indoor

15.4.2 Outdoor

15.5 Specialty Housings

15.5.1 High Security

15.5.2 Elevator

15.5.3 Dust-Proof and Explosion-Proof

15.5.4 Pressurized and Air- or Water-Cooled

15.6 NEMA Environmental Ratings

15.7 Housing Accessories

15.8 Housing Guidelines

15.9 Summary

15.1 OVERVIEW

There are many varieties of video camera housings and integrated cameras available for indoor and outdoor secu-rity applications. Standard shapes and forms they take include: (1) rectangular—mounted on a wall or ceiling,

· dome—mounted on a ceiling, wall, pole, and pylon,

· triangular—mounted in a corner, and (4) wedge— ceiling mounted. The two primary functions for these environmental housing are to protect the camera and lens from vandalism and the environment. To meet these

requirements, indoor and outdoor housings and inte-grated camera modules are fabricated from a variety of materials including aluminum, painted steel, stainless steel, and molded high-impact plastic.

There has been an increasing demand for aesthetically designed housings and cameras to match the decor of a building interior or exterior. While the primary function of the housing is to protect the camera, lens, and electri-cal wiring, these aesthetic camera housings are especially attractive and unobtrusive as dictated by architectural con-siderations. To satisfy these requirements, manufacturers have produced attractive designs using injection-molded plastic and other materials and forming techniques.

Housings are used to protect vital electronic video equipment; consequently, the material used for their construction must be chosen carefully. Underwriters Laboratories (UL) has developed guidelines for minimum fire-safety requirements and suggested tests and ratings for fireproof or fire-retardant designs. This is especially important for non-metallic designs. The Electronic Industries Association (EIA) has guidelines for improved interchangeability among manufacturers’ products. The National Electrical Manufacturers Association (NEMA) has detailed specifications describing the requirements for indoor and outdoor housing requirements of elec-trical equipment. These guidelines and ratings relate to materials and finishes, mechanical design parameters such as mounting-hole locations, and electrical-cable entry and fittings.

This chapter describes rectangular, triangular, dome, and all the other special indoor and outdoor housings, including accessories such as heaters, fans, thermostats, and windshield wipers and washers. Most housings have locks or tamperproof hardware to prevent vandalism or theft of the camera and lens.

387

388 CCTV Surveillance

Camera Housings. The indoor round hemispherical dome-shaped housing has become very popular because it is attractive and has excellent functionality. The dome’s symmetrical shape and tinted viewing “window” prevents the observer from seeing the direction in which the cam-era is pointing. This adds a deterrence factor to the surveil-lance function. Many security installations require discreet video surveillance equipment that blends in with the surrounding environment, not eye-catching or obtrusive housings. Corner-mounted triangular and wedge-shaped housings are also in widespread use.

Outdoor housings used on facility properties are designed to match landscaping and grounds and/or spe-cific lighting conditions. Outdoor environmental housings that are subject to wind loading or ice buildup should be no larger or heavier than required to house the camera, lens, and associated wiring and accessories. They should be constructed to withstand the harsh outdoor environment and added abuse from vandalism or attack. The camera housing enclosures should have easy access into them via a hinged or sliding interior assembly or removable cover.

The housing, camera, and lens are often within reach of personnel who could damage or remove the equipment. Of particular concern are high-risk locations such as jail cells, building exteriors, and public-access locations that require a more rugged housing fabricated from stainless steel or high-impact polycarbonate plastic. Figure 15-1 shows two examples of standard indoor and outdoor cam-era housings.

Integrated Cameras. With the increased use of video surveillance cameras, manufacturers, video integrators, and end-users have sought to simplify the purchasing and installation of camera systems. To that end the integrated camera has become very popular and an efficient means to accomplish that function. The integrated camera is a plug and play surveillance camera including the camera, lens, and any internal wiring associated with it, and mounted in a small housing that is ready to install at the site in a minimum amount of time. These integrated cameras take on shapes similar to some of the housings described in

the previous section but are smaller and more compact. Very popular types are domes, corner mount, wedge, with environmentally enclosed day/night camera with integral bracket mounting. Figure 15-2 shows examples of these integrated cameras.

15.2 INDOOR HOUSINGS

15.2.1 Functional Requirements

Indoor housings must protect the camera and lens from pollutants such as dust and other particulate matter, a cor-rosive atmosphere, and tampering or vandalism. Indoor housings are constructed of painted or anodized alu-minum, painted steel, stainless steel, and several types of plastic. The material for plastic housings must be flame-proof or flame-retardant, as designated by local codes and UL recommendations. The housings must have sufficient strength to protect the lens and camera, and be sturdily mounted onto a fixed wall or ceiling mount, or recessed in a wall or ceiling. The lens should view through a clear window made of safety glass or plastic. Recommended plastic window material is either high impact acrylic or polycarbonate with a mar-resistant finish. The electrical input/output access locations should be designed and positioned for easy maintenance. For easy access and servicing of internal parts, the top half of the housing should be hinged or be able to slide open, or be remov-able. In some designs, the entire camera/lens assembly is removable for servicing. Figure 15-3 shows the interior of a typical rectangular indoor housing.

The common rectangular housing is available in many sizes and is the least expensive. For vandalism protection, many housings are available with key locks or tamperproof hardware that allows the cover to be removed only with a special tool. In very high risk areas, welded stainless-steel housings with thick polycarbonate windows (3/8 or 1/2 inch) and high-security locks are used. Some housings

(A) INDOOR (B) OUTDOOR

FIGURE 15-1 Standard indoor and outdoor camera housings

Integrated Cameras, Camera Housings, and Accessories

389

(A) STAINLESS STEEL CORNER MOUNT (B) CEILING MOUNT-FIXED DOME

(C) SPRINKLER CEILING MOUNT (D) HARDENED WALL/CEILING

FIGURE 15-2 Popular integrated cameras

are designed to provide concealment and improved aes-thetics by recessing them into the wall or ceiling. The five housing types that account for most security installations are: (1) rectangular, (2) dome, (3) wedge, (4) triangular, and (5) wall- and ceiling-recessed and surface-mounted.

15.2.2 Indoor Types

Rectangular. The most popular type of housing is the standard rectangular design since it can be fabricated at low cost, is sturdy, and is available from many manufactur-ers in many sizes and attractive styles.

Under normal circumstances, indoor housings do not require any special corrosion-resistant finishes. The housings are made from painted or anodized alu-minum, painted steel, or high-impact plastic, such as polyvinyl chloride (PVC), acritile buterated styrene (ABS), or polycarbonate (General Electric Lexan, etc.). In high crime areas and jails, stainless steel housings are used.

Accessibility to the camera/lens assembly for installation and servicing is important. Video surveillance cameras are always mounted near or at ceiling height, on a pedestal, or at some elevated location requiring service personnel to be on ladders or power lifts. The housing design must permit

390 CCTV Surveillance

FIGURE 15-3 Indoor housing showing interior

easy access and serviceability under these conditions. Man-ufacturers provide one of several means to gain access to the housing: (1) removable top cover, (2) hinged top cover, (3) top cover or camera/lens on slide, (4) remov-able front and/or rear cover, (5) hinged bottom cover (dome), or (6) top cover on slide (Figure 15-4).

Dome. A second category of indoor housing is of a round or hemispherical, clear or tinted dome in which a cam-era, lens, and an optional pan/tilt mechanism are housed. Chapter 14 described dome cameras in detail. The ceiling-mounted hemispherical dome and the below-the-ceiling and wall-mounted domes on brackets look totally differ-ent from the rectangular housing, and often blend in better with architectural decor. Since they look like a lighting fixture, they are less obtrusive than rectangular housings. Since the hemispherical dome is circularly sym-metrical, it can be in a fixed position and the CCTV cam-era pointed in any direction to view the scene. A pan/tilt unit used in a dome can rotate and tilt the camera and lens while still remaining inside the confines of the dome. This is in contrast to cameras inside rectangular and other housings: if the camera moves, the entire housing assem-bly has to move as a unit.

If the dome is tinted so that the person down at floor level viewing the dome cannot see the camera and lens, it is possible to point the camera in any direction without the observer seeing it move. This capability can act as an additional security deterrent because the observer does not know when he or she is under surveillance.

There are three different types of plastic dome materials through which the lens views the scene: (1) clear, (2) semi-transparent aluminum- or chrome-coated, and (3) tinted or smoked plastic. When the dome housing is used for

REMOVABLE

TOP COVER

MAIN

MAIN

HOUSING

HOUSING

SLIDE

(1) REMOVABLE TOP COVER

(2) HINGED TOP COVER

(3) CAMERA/LENS SLIDE

REAR

COVER

MAIN

CEILIING LEVEL

SLIDE

HOUSING

HINGED

FRONT DOME

COVER

(4) REMOVABLE FRONT/REAR COVER (5) HINGED BOTTOM COVER (6) TOP COVER ON SLIDE

FIGURE 15-4 Camera housing access methods

protection only and its pointing direction need not be con-cealed, the clear plastic dome is the best choice, since it produces only a small 10 or 15% light loss. If the camera’s pointing direction is to be concealed for additional secu-rity a coated or tinted dome is required. The aluminized dome is the earliest version of the coated dome and atten-uates the light passing through it by approximately two f-stops (equivalent to approximately a 75% light reduction or loss). While this type of dome is still in use, the pre-ferred dome material is a smoked plastic or tinted plastic that attenuates the light approximately one f-stop, or 50%.

In contrast to rectangular housings using flat plas-tic or glass windows with excellent optical quality and transmission, some dome systems add slight optical dis-tortion to the video picture. In high-quality domes the image distortion is almost negligible, but in some systems the distortion or loss in resolution is noticeable. In any dome-housing application the camera/lens should view through the surface of the dome perpendicularly as shown in Figure 15-5a.

Under this condition, there is at least symmetry of dis-tortion, that is, the primary effect is that of a weak lens producing a small change in the focal length of the total

Integrated Cameras, Camera Housings, and Accessories

391

lensing system and is usually not noticeable. If the cam-era/lens pointing axis is not perpendicular to the dome surface (Figure 15-5b) and looks at an oblique angle through the dome housing material, noticeable distortion will occur; for example, images may appear elongated ver-tically or horizontally. If the dome and camera are in a fixed position with respect to one another, the distortion is generally less noticeable than if the lens is scanning or tilting while the dome remains still. Figure 15-6 shows four widely used dome housing configurations.

Wedge Housing. One version of the wedge housing is designed to replace an existing standard 2 feet × 2 feet drop ceiling tile (Figure 15-7a) and another version (Figure 15-7b) is designed for surface mounting. The wedge housing in Figure 15-7a is a manually rotatable 16-inch high impact white plastic center section with a wedge-shaped camera protruding about 5 inches below the ceiling line. There are no additional accessories required. The design allows for manual pan adjustments of 360 and minor tilt adjustments. After final pointing the center camera/lens section is restricted from rotating by tight-ening thumbscrews. The camera’s wedge shape aims the camera about 15 down from the horizontal. The front of

(A ) LENS AXIS PERPENDICULAR TO DOME SURFACE: EXCELLENT IMAGE

DOME

SURFACE

· LENS AXIS NOT PERPENDICULAR TO DOME SURFACE: POOR IMAGE

DOME

SURFACE

LENS AXIS

OBLIQUE

ANGLE

LENS AXIS

FIGURE 15-5 Indoor ceiling mounted dome camera with lens axis perpendicular to dome surface

392 CCTV Surveillance

(A) INDOOR-N CEILING MOUNT (B) OUTDOOR-BUILDING/POLE MOUNT

(C) INDOOR-SURFACE MOUNT (D) OUTDOOR-SURFACE MOUNT

FIGURE 15-6 Dome housing configurations

the protrusion has a viewing window of clear acrylic with no distortion and virtually no light transmission loss.
Another version is a small surface-mounted wedge-shaped housing that can be attached to any ceiling. These are available in either a surface- or recessed-mounting configuration.

Corner Mount. Figure 15-8 illustrates examples of aes-thetic and hardened camera/lens housings designed specifically for corner mounting in rooms, elevators, stair-wells, jail cells, etc. Figure 15-8a shows a high-security housing of welded stainless steel with a polycarbonate win-dow. The tamperproof corner mount camera housing has a camera bracket assembly permitting the camera to be tilted vertically ±10 for minor adjustments of the vertical pointing angle. The lens viewing window permits viewing a 95 horizontal FOV and 75 vertical FOV. The optimum pointing direction for the lens and camera is 45 with respect to both adjacent walls and 45 down from the ceiling horizontal plane. For an elevator cab application this housing with a wide-angle, 95 horizontal FOV can view entire elevator cab with no hidden areas and provide 100% video coverage of the cab area. The high-security housing has a hinged, lockable cover for easy, controlled access to all internal parts, and a tough mar-resistant poly-

carbonate (Lexan) window. All mounting, video, and elec-trical power access holes are located on the rear and top surfaces and inaccessible to the public. The installation meets codes that require unbroken firewalls. Three differ-ent housing sizes of this design accommodate most CCD solid-state cameras and wide-angle manual- or automatic-iris lenses or variable focus (vari-focal) lenses. Since the housing is exposed to the public, it is securely locked and manufactured using tamperproof materials, such as steel or stainless steel, and a polycarbonate (Lexan) window.

Figure 15-8b shows a housing fabricated from high impact plastic and is a configuration suitable for applica-tion requiring moderate security. The plastic housing has a lockable front cover and all mounting and electrical access holes are out of sight, and not accessible to the public. The housing has an adjustable bracket for tilting the camera vertically. There are many manufacturers supplying these types of corner mount housings in materials ranging from stainless steel, steel, and plastic. Finishes include brushed stainless steel and painted aluminum, steel, and plastic.

Figure 15-8c shows a mirror-view corner mount hous-ing that has a tinted or aluminized one-way window. It is

· small 7 inch × 7 inch × 7 inch unobtrusive housing that renders the camera and lens covert.

Integrated Cameras, Camera Housings, and Accessories

393

(A) ROTATABLE: 2′ × 2′ PANEL (B) FIXED HOUSING

(C) TYPICAL ABOVE CEILING HOUSING (D) COMPACT: 7″ LONG

FIGURE 15-7 Wedge camera housings

Ceiling- or Wall-Recessed or Surface Mount. Recessed or partially concealed housings are often mounted in ceil-ings and walls. Figure 15-9 shows examples of these hous-ings, including the wedge and dome-shaped types. The round, semicircular, and tapered housings shown offer design flexibility since the camera and lens can be pointed in any horizontal direction while the square or rectan-gular ceiling tile remains in place. These housings are used where a low-profile (but not covert) type of surveil-lance camera is required. These cameras are well suited for looking down hallways, at cash registers, etc. In ceiling installations, most of the housing, camera, and lens are mounted above the ceiling level. The only portion below ceiling level is a small part of the housing and the window through which the camera lens views. The cameras and lenses are accessible from below ceiling level by unlocking a cover that swings down, or by gaining access from the rear of the housing above the ceiling from an adjacent ceiling tile. It is important that all ceiling tile mount hous-ings be securely attached to a structural member of the

building above the ceiling with a chain or cable so that if the hanging ceiling support fails, the housing and con-tents do not fall to the floor or possibly injure personnel below.

With the increased use of video surveillance in pub-lic locations, be they government, industrial, or private, more attention is being given to the decorative and aes-thetic features of the housing. These housings often have finishes of brass, gold, or chrome, with satin or polished finishes. They are also available with custom paint colors and textures, and custom-colored plastics. Several manu-facturers offer special shapes and custom configurations for matching specific architectural designs.

15.3 OUTDOOR HOUSINGS

Like the indoor housing, the outdoor housing protects the camera and lens from vandalism and adverse out-door environments. Most outdoor housings are provided

394 CCTV Surveillance

(A) DISCRETE TRIANGULAR ONE-WAY MIRROR (B) DISCRETE CONVEX TINTED MIRROR

(C) STAINLESS STEEL (D) HIGH IMPACT PLASTIC

FIGURE 15-8 Corner mount housings

with key locks to prevent unauthorized opening of the housing.

15.3.1 Functional Requirements

Outdoor housings must protect the camera from vandal-ism as well as adverse environmental conditions. The van-dalism encountered can range from rocks or sticks thrown at the housing to bullets and other explosives. These secu-rity housings are prime targets since they are mounted on ceilings, walls, building exteriors, and poles and pedestals.

In outdoor installations the camera is mounted in a protective enclosure to protect it against environmental factors such as precipitation: rain, hail, snow, sleet, ice, and condensing humidity. The outdoor housing must also protect against many types of particulate matter including

dirt and dust, sand, fly ash, soot, and any other material local to a particular site. Outdoor locations with a cor-rosive atmosphere can cause rapid deterioration, failure, and premature replacement of the camera and lens if not properly protected. These substances include industrial chemicals, acids, and salt spray. Outdoor housings should have external finishes that withstand the atmosphere in which they are to operate. In hot climates, a sun shield or shroud and a bright aluminum or white finish is desir-able to reflect sunlight and eliminate heat buildup in the housing.

Outdoor housings share many of the same require-ments as indoor housings. Accessibility to the camera and lens during installation and maintenance are more impor-tant in outdoor applications since video equipment is often mounted high above the ground and serviced under adverse conditions.

Integrated Cameras, Camera Housings, and Accessories

395

(A) CONCEALED CEILING (B) WEDGE

(C) DOME

FIGURE 15-9 Recessed and concealed ceiling, wedge, and dome housings

15.3.2 Outdoor Design Materials

Outdoor housings are manufactured from aluminum, painted steel, stainless steel, and outdoor-rated plas-tic, including polycarbonate, ABS with a UV protective layer. It is important that plastic outdoor housings be fabricated from UV-inhibiting materials, to prevent the housing from deteriorating due to sunlight. Plastics not treated will crack, and colors will fade. High-quality baked-enamel, painted-steel, and stainless-steel housings will

15.3.3 Outdoor Types

The outdoor camera housings are similar to the indoor except that they must be furnished with an exterior finish that can resist and withstand the outdoor environment. They should be fitted with a thermostatically controlled heater and fan so that when the temperature extends beyond the range of the camera and lens specifications they can either be heated or cooled.

last many years. Where long-lasting, high-security, vandal-

Rectangular. For outdoor applications the rectangular

proof housings are required, stainless steel is the choice

plastic, painted aluminum, or stainless steel housings

since it does not rust or corrode and is extremely tough.

are the most popular choices. These housings are eas-

Aluminum is a good choice for an outdoor hous-

ily mounted from a bracket on a building, wall, or pole,

ing when anodized and finished in baked polyurethane

or hung from a building overhang to provide a solid

enamel paint and anodized. Anodized and painted alu-

mounting.

minum is the most durable finish. Aluminum and steel

housings should not be used when a salt or other cor-

Dome. Dome housings can be mounted on an individual

rosive atmosphere is expected. Stainless steel and special

pole or pylon, under the eaves of a building, or on a

plastics are the best choice for a salt-spray environment.

bracket mounted off the wall of a building. These housings

Consult the housing or materials manufacturer for the

must also use outdoor materials that will withstand the

proper choice.

environment.

396 CCTV Surveillance

15.4 INTEGRATED CAMERAS

The integration of the video camera, lens, housing, and mount into one unit has been a natural evolution in the security industry. This evolution has occurred as a result of the availability of small CCD and CMOS cameras and asso-ciated small lenses. It has made technologic and economic sense for manufacturers to integrate these components into a single finished product available to the video sys-tems integrator or end-user as a plug and play video surveil-lance module ready for mounting on a wall, a ceiling, outside a building, etc. These integrated cameras have taken the form of domes (see Chapter 14), triangular-corner, wedge, and covert. There are many manufacturers producing hundreds of models for indoor and outdoor applications. They are available in monochrome and color for daylight and nighttime use.

15.4.1 Indoor

Indoor integrated cameras have housings that take the form of those described in Section 15.2. The housing types

include the dome, triangular-corner, wedge, and semi-covert models (Figure 15-10).

Dome. The integrated dome camera uses a dome hous-ing with a camera and lens installed. Most dome applica-tions now use the integrated dome camera instead of the component form because of the ease of installing a com-plete plug and play module and the concomitant lower overall cost. These modules are available for monochrome and color use as well as total darkness using infrared LED illumination. Figure 15-11 illustrates an integrated dome camera and its interior assembly in an electrical duplex outlet box showing the manually adjustable and tilt bracket for the camera lens assembly.

Triangular-Corner. A triangular-shaped integrated cam-era housing using a one-way mirror installed in the corner of a room at the ceiling level provides an excellent semi-covert surveillance camera. Typical locations are in a small room or lobby, an elevator or a stairwell. Figure 15-12 shows this design using a wide-angle (90 FOV) lens that can view the entire area of a small room or other space.

(A) DAY/NIGHT RUGGEDIZED DOME (B) SPRINKLER HEAD

(C) CORNER MOUNT MIRROR (D) RUGGEDIZED WALL MOUNT

WITH LED IR ILLUMINATION

FIGURE 15-10 Indoor integrated cameras

Integrated Cameras, Camera Housings, and Accessories

397

FIGURE 15-11 Integrated dome camera assembly FIGURE 15-13 Wedge integrated camera

15.4.2 Outdoor

FIGURE 15-12 Discrete triangular corner mount mirror integrated camera

The camera installed in the triangular housing is at a 45 angle pointing down from the ceiling to view the entire area. The triangular housing can be mounted in protected outdoor locations at entrances or exits to buildings, etc. where two walls meet, resulting in a very unobtrusive instal-lation. When mounted in hot or cold environments, the housings must be provided with a thermostatically con-trolled heater or fan.

Wedge. The wedge-integrated camera is available as a small, unobtrusive assembly suitable for mounting directly to a hard ceiling or on a ceiling tile. These cameras are lightweight and generally require no additional support structure—they can be mounted directly onto the ceiling tile (Figure 15-13).

Covert. There are many variations of integrated covert-type video surveillance cameras used to augment overt cameras. These can take the form of a sprinkler head, smoke detector, passive infrared detector, temperature thermostat, etc. (see Chapter 18 for many versions of covert integrated cameras).

Most integrated camera units for outdoor applications take the form of a dome camera assembly in a plug and play form for maximum ease of installation and servicing. Some other forms used include ruggedized camera hous-ings with the camera, lens, heater, and fan, all enclosed in the housing, ready for mounting on an exterior bracket pole or pedestal. Dome assemblies such as those shown in Chapter 14 for outdoor applications are representative of these types.

15.5 SPECIALTY HOUSINGS

There are security applications in which cameras must be located in very hostile environments. To protect the camera and lens from damage and downtime, manufac-turers offer housings that can withstand high mechanical impact from hand-thrown or fired projectiles, extreme high temperature, dust, sand, liquid, corrosive chemicals, and explosive gas. The following housings have unique characteristics for solving these extreme security or special environmental applications.

15.5.1 High Security

There are numerous armored camera/lens enclosures for installation in correctional institutions. Figure 15-14 illustrates several high-security housings designed specif-ically for mounting in jails and detention and holding cells, to provide maximum protection from vandalism. These integrated cameras have no exposed hardware or cabling and all use heavy-duty high security locks with tamper switches. The housings are fabricated from 10-gauge (0.134-inch thick) or heavier welded steel. The win-dow material is 3/8–1/2-inch polycarbonate or cast acrylic plastic having an abrasion-resistant finish. These housings withstand blows and impacts from hammers. Rocks and

398 CCTV Surveillance

FIGURE 15-14 High security integrated cameras

(A) CEILING MOUNT (B) IN-WALL

(C) CORNER MOUNT (D) WALL MOUNT

some firearm projectiles cannot penetrate or destroy the integrity of the housing.

15.5.2 Elevator

Figure 15-15 illustrates an example of a hardened cam-era/lens housing designed specifically for elevator appli-cations. The photograph of the elevator interior illustrates that the full interior of an elevator can be monitored from one wide-angle camera/lens system.

The elevator housing style is available in three sizes: 6, 8, and 12 inches high. These high-security housings are fabri-cated from welded stainless steel with a 1/4-inch thick poly-carbonate window. The tamperproof integrated camera assembly is complete with a monochrome or color CCD camera and a wide-angle, 90 FOV lens in the stainless-steel housing. In this configuration, the camera can be tilted

±10 for minor adjustments of the vertical pointing direc-tion. The high-security housing has a hinged, lockable cover for easy, controlled access to all internal parts, and

a tough mar-resistant polycarbonate (Lexan) viewing win-dow. All mounting and camera power and video electrical cable access holes are located on the rear and top surfaces, and are inaccessible to the public. The installation meets codes that require unbroken firewalls. The three housing sizes accommodate most CCD solid-state cameras using wide-angle manual- or automatic-iris or vari-focal lenses. These integrated cameras can also accommodate cameras with infrared LED lighting to obtain excellent viewing under completely dark, unlighted conditions. The small 6-inch high unit accommodates and protects all small 1/4-and 1/3-inch format cameras and associated wide-angle lenses. Figure 15-16 illustrates the camera viewing and pointing parameters for elevator-cab surveillance.

The optimum pointing direction for the lens and cam-era is 45 with respect to both adjacent walls and 45 down from the ceiling horizontal plane. With a wide-angle, 90 horizontal FOV the entire elevator cab is viewed with no hidden areas, providing 100% video coverage of the cab area. Since the housing is exposed to the public, it is securely locked and is manufactured using tamperproof steel and stainless steel, and a polycarbonate window.

Integrated Cameras, Camera Housings, and Accessories

399

(A) STAINLESS STEEL HOUSING (B) CAMERA VIEW

FIGURE 15-15 High security corner mount elevator integrated camera

CEILING

WALL WALL

VERTICAL

FOV = 75°

HORIZONTAL

FOV = 95°

LENS VERTICAL POINTING

DIRECTION: 45°

LENS HORIZONTAL POINTING FLOOR
DIRECTION: 45°

FIGURE 15-16 Elevator cab viewing parameters

15.5.3 Dust-Proof and Explosion-Proof

The dust-proof housing is similar to many other camera housings except that it is totally sealed from the out-side atmosphere and therefore can be used in sandy and dusty environments (Figure 15-17). When fabricated from stainless steel, these housings can withstand the effects of corrosive environments. The window material is tem-pered glass to provide safety and maximum resistance to abrasion and corrosion. To provide some cooling of the camera and lens, a fan is used to circulate the air inside the housing, and an optional sun shield above the camera housing protects it from direct solar heating. The housing is provided with air fittings so that an external, filtered,

compressed-air supply can be used to maintain moderate operating temperatures. These housings are not consid-ered explosion-proof.

Explosion-proof housings are designed to meet the rig-orous safety requirements of explosion-proof and dust-ignition-proof electrical equipment, for installation and use in hazardous locations (Figure 15-18). These security housings and cameras meet the requirements of the National Electric Code Class 1, Division 1, and Class 2, Division 1, and are certified as per the require-ments of UL 1203 specifications and procedures. These housings are generally of heavy-wall, all-aluminum con-struction and are available in 6, 8, and 10 inch diam-eters to accommodate most camera/lens combinations.

400 CCTV Surveillance

FIGURE 15-17 Dust-proof integrated housing and camera assembly

FIGURE 15-18 Explosion-proof integrated housing and camera assembly

They are fitted with explosion-proof, sealable fittings for electrical power/control input and video signal output. Optional sun shrouds are available for operation in hot environments.

15.5.4 Pressurized and Air- or Water-Cooled

Pressurized housings are used in hazardous atmospheres. They meet these requirements by purging (filling) them with an inert gas at a pressure in accordance with National Fire Protection Association specification Number 946 (Figure 15-19).

The housings are fabricated from thick-walled alu-minum with corrosion-resistant finishes. The window is 1/2-inch-thick tempered and polished plate glass. These housings can be back-filled (purged) with low-pressure nitrogen gas to a pressure of 15 pounds per square inch gage (psig). Nitrogen is completely inert and prevents an explosion from occurring if there is any spark or electrical malfunction in the housing. The housings have hermeti-cally sealed O-ring seals located between the access cover and the housing. All electrical terminations are made and brought out through hermetic seals. To purge the housing, the access cover is mounted and secured, and the housing is filled with dry nitrogen to a pressure of 15 psig by means of a filling valve and pressure-relief valve. The purge is then closed and the nitrogen filling tube removed. These housings are significantly more expensive than standard housings, since they must be designed to be

(A) PRESSURIZED OUTDOOR DOME (B) PRESSURIZED AND NITROGEN PURGED (C) WATER COOLED

FIGURE 15-19 Pressurized and water cooled environmental housings

Integrated Cameras, Camera Housings, and Accessories

401

hermetically sealed to provide a positive pressure of 15 psig differential pressure, and to withstand an explosion.

Water-cooled housings are designed for use in extremely hot indoor or outdoor locations. They require a constant supply of cooling water for proper operation. A 1-inch-thick water jacket built into the housing effectively shields the camera/lens from the outside environment. Depending on the application, the housings are made

on these ratings is included on the manufacturer’s liter-ature, and detail information can be obtained from the NEMA organization. Table 15-1 summarizes several NEMA housing ratings for indoor and outdoor designs.

15.7 HOUSING ACCESSORIES

from aluminum or stainless steel. An internal fan pro-

There are numerous accessories available for indoor and

outdoor housings. Some of the more common types

vides constant air circulation within the housing, aids in

include thermostatically controlled heaters and fans, win-

efficient heat transfer to the water jacket, and prevents

dow wipers and washers, sun shields and shrouds, and

heat buildup. The housing is supplied with a 1/4-inch-

many types of mounts and brackets.

thick Pyrex heat-resistant window for operating at temper-

atures up to 550 F (288 C). Consult the manufacturer to obtain recommendations for the specific operating environment.

15.6 NEMA ENVIRONMENTAL RATINGS

The NEMA has developed a comprehensive set of specifi-cations and ratings for indoor and outdoor electrical hous-ings. Many of the manufacturers of video security housings and integrated camera modules have designed their prod-ucts to meet some of these housing ratings. Information

Heater and Fan. In warmer climates where the tem-perature does not drop below freezing, only a fan and thermostat are required to maintain suitable operating temperatures for the camera and lens. The thermostat is designed to automatically turn on the fan when the tem-perature in the interior of the housing rises above some value, usually between 90 and 100 F (32–38 C), and turn it off when it falls a few degrees below the set tempera-ture. In cold climates, a heater and thermostat are used to keep the lens and camera above about 45–55 F (7–13 C). The heater prevents condensation on the window and lens and keeps the automatic-iris mechanism and cam-era operative. In freezing weather, it prevents moisture

PROVIDES PROTECTION AGAINST THE

FOLLOWING ENVIRONMENTAL CONDITIONS

APPROXIMATE IP EQUIVALENT **

INCIDENTAL CONTACT WITH ENCLOSED EQUIPMENT

INDOOR

OUTDOOR

FALLING DIRT

DRIPPING AND LIGHT SPLASHING LIQUIDS

RAIN, SLEET AND SNOW

CIRCULATING DUST, LINT, FIBERS, DEBRIS

SETTLING DUST, LINT, FIBERS, DEBRIS

EXTERNAL ICE

HOSEDOWN AND SPLASHING WATER OIL AND COOLANT SEEPAGE
OIL AND COOLANT SPRAYING AND SPLASHING CORROSIVE AGENTS

OCCASIONAL TEMPORARY SUBMERSION

OCCASIONAL PROLONGED SUBMERSION

· NEMA—NATIONAL ELECTRICAL MANUFACTURERS ASSOCIATION

· IP—INGRESS PROTECTION CLASSIFICATION

4 AND 4X ARE THE MOST COMMONLY USED OUTDOOR TYPES 12 AND 13 ARE THE MOST COMMONLY USED INDOOR TYPES

NEMA ENCLOSURE TYPE *

1

3

4

4X

6

6P

12

13

IP30

IP64

IP66

IP66

IP65

IP65

Table 15-1 NEMA Housing Ratings for Non-Hazardous Locations

402 CCTV Surveillance

from freezing on the window and within the environmen-tally enclosed housing. The thermostat applies power to the heater when the temperature goes down below the dew point. When the camera/lens housing is located in an interior close-to-the-ceiling environment or in an out-door warm environment, a thermostatically controlled fan is used to cool the camera/lens combination. The fan should contain a removable filter that can be cleaned or replaced periodically.

Heaters require considerable electrical power for their operation. Most heater assemblies supplied by the man-ufacturer require 24 VAC for their operation. If primary power is supplied from a 117 VAC source then a step-down transformer with a 24 VAC output is required. If 117 VAC power is not locally available at the camera-housing site, the wire supplying the power must be sized correctly. Table 15-2 lists appropriate wire sizes vs. distance between the 117 and 24 VAC sources and the camera.

Window Washer and Wiper. Another accessory is the win-dow washer and wiper. If the housing is rectangular and pointing down at 15 or 20 or more, it is generally unnec-essary to provide the housing with a window wiper and washer, as rain will run off the window, along with dirt, and allow proper viewing. If, however, the housing is located in a dusty environment or is in a more horizontal direction, it is advisable to include a window washer/wiper assem-bly. This assembly is mounted below and in front of the

window and operates like an automobile washer/wiper sys-tem. The wiper motor and liquid washing pump can be energized automatically and periodically or remotely from the control console.

Most environmental housings, indoor or outdoor, are supplied with plastic or safety (tempered) glass windows for the lens to view through. These windows may be acrylic, polycarbonate, or glass, depending on the design. The choice of acrylic vs. polycarbonate depends on whether the application is to be maximally tamperproof or only mod-erately so, and whether the housing is used indoors or out-doors. Acrylic is optically clear and will transmit over 95% of the light. Polycarbonate transmits less—approximately 85%—but has a higher impact resistance than acrylic. Both types are available in a mar-resistant type which is highly recommended, and will remain optically clear under nor-mal cleaning action and withstand outdoor weathering. For maximum resistance to scratching and abrasion, safety glass is used. Window thicknesses range from 1/8 inch for light duty to 1/4 inch for normal service and from 3/8-to 1/2-inch for maximum security housings. For dome systems the portion of the dome that is used for viewing has its surface pointing downward and tends to be self-cleaning; however, they must be cleaned periodically and water droplets on the surface will reduce visibility.

Tamper Switch. In most security applications, it is impor-tant that when the camera housing is being opened by authorized or unauthorized personnel, the system or

POWER

CONDUCTOR

RESISTANCE

POWER TO HEATER AND CAMERA OVER TWO CONDUCTOR CABLE

MAXIMUM CABLE LENGTH (ft)

SOURCE

SIZE

ohms/1000 ft †

VOLTAGE

AWG *

25 WATT LOAD (0.21 AMP)

50 WATT LOAD (0.42 AMP)

100 WATT LOAD (0.84 AMP)

22

33.0

1656

828

414

20

20.8

2628

1314

657

117 VAC

18

13.02

4198

2099

1050

16

8.18

6683

3341

1671

14

5.16

10594

5297

2649

12

3.24

16872

8398

4199

25 WATT LOAD (1.04 AMP)

50 WATT LOAD (2.08 AMP)

100 WATT LOAD (4.16 AMP)

22

33.0

69

34

17.3

20

20.8

110

55

27.5

18

13.02

176

88

44

24 VAC

16

8.18

281

140.5

70

14

5.16

445

222.5

111

12

3.24

709

354.5

177

10

2.04

1127

563.5

284

POWER

SOURCE

CAMERA

HEATER, ETC.

117 VAC, 24 VAC, 12 VDC ††

· AMERICA WIRE GUAGE
· RESISTANCE REPRESENTS FULL WIRE LENGTH, I.E. 2x CABLE LENGTH

· IF 12 VDC POWER IS USED, USE THE 24 VAC TABLE ABOVE AND DOUBLE THE WIRE LENGTH Note: BASED ON MAXIMUM VOLTAGE DROP OF 10%

Table 15-2 Wire Size vs. Distance for Housing Heater, Camera, and Other Electronics

Integrated Cameras, Camera Housings, and Accessories

403

(A) WALL (B) CEILING (C) OUTDOOR DOME MOUNT

FIGURE 15-20 Camera housing brackets and mounts

guard be alerted. An electrical switch in the camera hous-ing is used to activate an electrical alarm that can be sent back to the monitoring location when the housing has been opened.

Locks, Security Screws. There are various levels of secu-rity key locks available for indoor and outdoor housings. Most camera housings are supplied with standard locks, but these can be upgraded to high security locks when the application demands it. In place of key locks various types of security screw hardware is available. The manufacturer should be consulted on the different types of key locks and security screws that can be supplied.

Brackets and Mounts. A large variety of brackets and mounts are available to mount cameras, housings, and pan/tilt platforms safely to walls, ceilings, poles, pedestals, and other structures. Since most mounts are not compati-ble from manufacturer to manufacturer, the housing and bracket should be purchased from the same manufacturer to avoid extra costs for reworking parts that do not inter-face properly. Figure 15-20 shows some common camera housing brackets and mounts available.

15.8 HOUSING GUIDELINES

The EIA has written a guideline of recommended design parameters for housing manufacturers for hole configu-rations on mounting brackets and housing mountings. At present, not all manufacturers use the same mounting-hole configuration. The EIA has recommended guidelines

for the electrical input/output wiring and connector con-figurations so that there is interchangeability between manufacturers and so that safe procedures are followed by manufacturers and installers. Local building codes and UL codes specify the minimum requirement for electrical enclosure materials. They should be consulted to be sure materials are suitable. The purchaser must be aware of the requirements for each application and look carefully at the manufacturer’s specifications to determine the most suitable housing. The NEMA housing recommendations should be consulted to help determine the specific rating for indoor or outdoor housings.

15.9 SUMMARY

The security camera housing plays an important role in protecting the camera and lens from the environment and vandalism, and insuring that they will be in a safe and controlled environment to maximize life and picture quality. Many camera housing designs are available for indoor and outdoor applications.

In an effort to reduce the complexity of choosing a com-patible lens, camera, and other accessories at the camera site, the integrated camera design has evolved. This inte-grated design is lower in cost and requires less installation time resulting in an additional cost savings.

There are many specialty housings to protect the camera and lens in harsh environments and from extreme vandal-ism. With the large number of housing manufacturers to choose from, there is a housing configuration for almost any application.

Chapter 18

Covert Video Surveillance

CONTENTS

18.1 Overview

18.2 Covert Techniques—Background

18.3 Covert Lens/Camera Types

18.3.1 Pinhole Lenses

18.3.2 Convertible Pinhole Lens Kit

18.3.3 Mini-Lenses

18.3.3.1 Off-Axis Optics

18.3.3.2 Optical Attenuation Techniques

18.3.3.3 Mini-Camera/Mini-Lens

Combination

18.3.4 Comparison of Pinhole Lens and

Mini-Lens

18.3.5 Sprinkler-Head Pinhole Lenses

18.3.6 Mirror-Pinhole Lens

18.3.7 Fiber-Optic Lenses

18.3.7.1 Configuration

18.3.7.2 Rigid Fiber Pinhole Lens

18.3.7.3 Flexible Fiber

18.3.7.4 Image Quality

18.3.8 Bore-Scope Lenses

18.4 Special Covert Cameras

18.4.1 PC-Board Cameras

18.4.2 Remote-Head Cameras

18.5 Infrared Covert Lighting

18.5.1 Concealment Means

18.5.2 IR Sources

18.6 Low-light-level Cameras

18.7 Imbeded Covert Camera Configurations

18.8 Wireless Transmission

18.9 Covert Checklist

18.10 Summary

18.1 OVERVIEW

Overt video surveillance equipment is installed in full view of the public and is used to observe personnel and activity

and letting people know that they are under surveillance. Overt video has had the effect of deterring crime of all types. Covert video ideally operates so that the offender is not aware of the surveillance. It can be recorded to produce a permanent video recording for later use in con-fronting, dismissing, or prosecuting the offender. Overt video security installations are very useful in apprehend-ing offenders; however, in special situations, investigators, police officials, government agencies, retail operations, and security personnel require covert or hidden cameras.

Covert and overt video are often used together to foil professional criminals. The criminal, seeing the overt sys-tem, defeats or disables the overt cameras, but the covert cameras can still record the activity. An unrelated reason for using covert video is to avoid changing the architec-tural aesthetics of a building or surrounding area.

Covert video cameras and lenses have become com-monplace, and although these hidden cameras use small optics, they can produce high-quality video images. Covert video cameras are concealed in common objects or located behind a small hole in an opaque barrier (such as a wall or ceiling). Cameras are camouflaged in common objects such as lamps and lamp fixtures, table and wall clocks, radios, or books. A very effective covert system uses a camera and lens camouflaged in a ceiling-mounted sprin-kler head.

This chapter will analyze covert video principles, tech-niques, and unique pinhole lenses and cameras. Lenses are analyzed that have a small front lens diameter thereby permitting the lens and camera to view the scene through a 1/16-inch-diameter hole. Most of these lenses have a medium-to-wide FOV, from 12 to 78 , to cover a large scene area, but still permit identification of persons and the monitoring of activities and actions. Special pin-hole lens variations including right-angle, automatic-iris, sprinkler-head, and fiber optic are described, as well as small pinhole cameras combining a mini-lens and sensor

445

446 CCTV Surveillance

into a small camera head and other complete minia-ture cameras.
In low-light-level (LLL) applications, a CCD camera with a very sensitive sensor and IR light source or an image intensifier is used. Since many covert installations are tem-porary, wireless transmission systems are used to send the camera signal to the monitor, recorder, or video printer.

18.2 COVERT TECHNIQUES—BACKGROUND

The lens and camera concealment is accomplished by having the lens view through a small hole, a series of small holes, or from behind a semitransparent window. Figure 18-1 shows a typical room in which covert video surveillance is installed.

A number of suitable covert camera locations include the ceiling, a wall, a lamp fixture, a clock, or other articles normally found in the room. Video cameras are installed in one or more locations in the room depending on the activity expected. Covert video systems using small lenses pose unique optical problems compared with overt systems that use standard lenses. Since the diameter of the front lens that views the scene must, by necessity, be small in order to be hidden, the lens is designed to be

optically fast, collecting and transmitting as much light as possible from the reflected scene to the camera sen-sor. As a consequence, small-diameter lenses called pin-hole lenses are used. (The term pinhole is a misnomer, as these lenses have a front diameter anywhere from 1/16 to 3/8 inch.)

There are several misconceptions regarding the factors determining a good pinhole camera or lens system for covert applications. Figure 18-2 shows the covert security problem. The lens/camera must receive reflected light from an illuminated scene. The lens must collect and transmit the light to the camera sensor and the camera must transmit the video signal to a remote video moni-tor and/or recorder and video printer. Most covert pin-hole lenses are designed for 1/4- and 1/3-inch camera sensor formats. For indoor applications the light sources are typically fluorescent, metal-arc, mercury, or tungsten types. Outdoor light sources include sunlight in the day-time, and mercury, metal-arc, tungsten, sodium, or xenon lighting at night. Figure 18-3 shows two basic configura-tions for pinhole lenses and cameras located behind a barrier.

The hole in the barrier is usually chosen to be the same diameter (d) or smaller than the pinhole lens front lens element. When space permits the straight-type

FIGURE 18-1 Covert CCTV lens/camera environment

Covert Video Surveillance

447

ILLUMINATION

SOURCE

SCENE

SMALL HOLE

LENS FOV

IN WALL

COVERT

PINHOLE LENS
ROOM AND CAMERA

BARRIER

INTEGRAL SMALL CAMERA

AND PINHOLE LENS

FIGURE 18-2 Covert CCTV surveillance

installation is used. In confined or restricted locations with limited depth behind the barrier, the right-angle pinhole lens/camera is used. In both cases, to obtain the full lens FOV it is imperative that the pinhole lens front lens ele-ment be located as close to the front of the barrier as possible to avoid “tunneling” (vignetting). When the pin-hole lens front lens element is set back from the barrier surface, the lens is, in effect, viewing through a tunnel, and the image has a narrower FOV than the lens is capable of producing. This appears on the monitor as a porthole-like (vignetted) picture.

An important installation problem often initially over-looked is the lens pointing angle required to see the desired FOV (Figure 18-4). Many applications require that the lens/camera point down at a shallow depression angle (30 ) from the ceiling (Figure 18-4a). This is accomplished by using the small-barrel, slow-taper lens. This feature allows pointing the small-barrel lens over a larger part of a room than the wide-barrel lens. Not all lenses can be mounted at a small angle to the ceiling because of the lens barrel shape (Figure 18-4b). Lenses having a large barrel diameter and fast taper at the front cannot be mounted at the shallow angles required. The small-barrel, slow-taper design permits easier installation than the fast-taper since less material must be removed from the barrier, and the

lens has a faster optical speed, since the front lens element is larger and collects more light. Figure 18-4 illustrates this installation problem. It shows a small hole on the scene side of the barrier and some material cut out of the barrier behind it to permit the front lens element to be located close to the front of the barrier surface. A pinhole lens having a small front diameter is simple to install. The smaller tapered barrel can be mounted at a smaller angle to the barrier than the wide-barrel lens. This feature allows pointing the small-barrel lens over a larger part of a room than the wide-barrel lens.

18.3 COVERT LENS/CAMERA TYPES

Pinhole lenses and cameras used for covert security appli-cations include: standard pinhole, compact pinhole lens kit, and mini-lens. There are many single board covert camera designs available using a small lens mounted to a single printed circuit (PC) board housed in a plastic or metal housing (Section 18.4.1). Special covert lens and camera designs include: fiber optic, sprinkler-head, and covert camera/lens combinations uniquely configured in special housings.

448 CCTV Surveillance

LIGHT SOURCE:

SUN

DETAIL OF

LAMPS: •

FLOURESCENT

PINHOLE LENS

TUNGSTEN (HALOGEN)

INSTALLED IN

SODIUM

ROOM BARRIER

MERCURY

INFRARED

REFLECTED LIGHT

FROM SCENE

SMALL

HOLE IN

WALL

SCENE

STRAIGHT

PINHOLE LENS

AND CAMERA

ROOM

BARRIER

RIGHT ANGLE

PINHOLE LENS

AND CAMERA

MONITOR

FIGURE 18-3 Straight and right angle pinhole installation

18.3.1 Pinhole Lenses

Figure 18-3 shows how pinhole lenses and cameras are mounted behind a wall, with the lens viewing through a small hole in the wall. Most are designed for 1/4 -, and 1/3-inch format cameras and have a manual- or automatic-iris control to adjust the light level reaching the camera. Figure 18-5 shows several samples of the generic pinhole lens types available.

The right-angle version permits locating the camera and lens inside a narrow wall or above a ceiling. The optical speed or f-number (f/#) of the pinhole lens is important for the successful implementation of a covert camera sys-tem. The lower the f-number, the more light reaching the camera and the better the video picture. The best theoret-ical f-number is equal to the lens focal length (FL) divided by its entrance lens diameter (d):

f /# = FL/d

(18-1)

This theoretical f-number cannot be obtained in practice because of various losses caused by imperfect lens trans-mission that is caused by reflection, absorption, and other lens-imaging properties. The light getting through the

pinhole lens to the camera sensor is limited primarily by the diameter of the front lens or the mechanical open-ing through which it views. The larger the lens entrance diameter, the more light getting through to the camera sensor, resulting in better picture quality, all other condi-tions remaining the same. The light collected and trans-mitted through a lens system varies inversely as the square of the lens f-number. If the lens diameter is increased (or decreased) a small amount, the light passing through the lens increased (or decreases) by a large amount: if the lens diameter is doubled, the light throughput quadru-ples. An f/2.0 lens transmits four times as much light as an f/4.0 lens. The f-number relationship is analogous to water flowing through a pipe: if the pipe diameter is dou-bled four times as much water flows through it. Likewise if the f-number is halved, four times as much light will be transmitted through the lens.

Many types of covert lenses are commercially available for video surveillance applications. Table 18-1 summarizes the characteristics of most manual- and automatic-iris pin-hole lenses.

Most of these lenses are designed for 1/4 – and 1/3-inch sensor formats since covert cameras are small. In spite of their small size they have resolutions of 380–420 TV

Covert Video Surveillance

449

(A) SLOW-TAPER BARREL

(B) FAST-TAPER BARREL

SMALL

LARGE

DIAMETER

DIAMETER

30°

55°

FIGURE 18-4 Pinhole lens pointing angle

SLOW TAPER FAST TAPER

STRAIGHT
STRAIGHT
MANUAL IRIS
MANUAL IRIS

RIGHT ANGLE

MANUAL IRIS

STRAIGHT

RIGHT ANGLE AUTO IRIS
AUTO IRIS

FIGURE 18-5 Standard straight and right-angle Pinhole lenses

450 CCTV Surveillance

ANGULAR FIELD OF VIEW (FOV) IN DEGREES

FOCAL

CAMERA FORMAT (inch)

TYPE

LENGTH

f/#

1/4

1/3

1/2

(mm)

HORIZ

VERT

HORIZ

VERT HORIZ

VERT

2.6

2.5

62.4

46.8

83.2

62.4

STRAIGHT

4.0

2.0

51.2

38.4

68.5

43.1

STRAIGHT

5.5

3.0

30.2

23.6

38.7

31.0

60.4

47.2

STRAIGHT

6.2

2.0

32,4

24.3

42.8

30.1

56.1

42.1

STRAIGHT

8.0

2.0

21.8

16.7

29.4

22.0

43.6

33.4

STRAIGHT

8.0

2.2

21.8

16.7

29.4

22.0

43.6

33.4

RIGHT-ANGLE

9.0

3.4

22.3

16.8

29.5

22.1

39.1

29.3

STRAIGHT

11.0

2.3

16.2

12.3

21.6

16.1

32.4

24.6

STRAIGHT

11.0

2.5

16.2

12.3

21.6

16.1

32.4

24.6

RIGHT-ANGLE

Table 18-1 Covert Pinhole Lens Parameters

MOUNT

CS

CS

C/CS

CS

C/CS

C/CS

C/CS

C/CS

C/CS

lines for a 1/4 – or 1/3-inch color camera and 450–570 TV lines for monochrome cameras. Many pinhole lenses have very small entrance apertures: 0.10 inch (2.5 mm) and are therefore optically slow (f/3.5–f/4.0) by design. From Equation 18-1 a lens with a FL of 9 mm and a 2.5 mm aperture (d) has at best a theoretical f-number of:

f/# = 9 mm/2.5 mm = 3.6

(18-2)

Other lens losses within this type of lens give an overall optical speed of approximately f/4.0.

A covert lens with an 11 mm FL and a 6 mm aperture has a theoretical f-number of:

f/# = 11 mm/6 mm = 1.83

(18-3)

Other lens losses result in an overall optical speed of approximately f/2.0. This means that the 11 mm lens col-lects four times as much light as the 9 mm lens.

The 9 mm lens with the smaller aperture works well if there is sufficient light. An advantage of the 6 mm-aperture (approximately 0.25 inch) lens is that it can be used in applications where a larger hole, that is, 6 mm diameter adequately conceals the lens and there is insufficient light available for the 9 mm FL lens with the 2.5 mm hole. The most important characteristics of a pinhole lens are: (1) how fast is the lens optical speed—that is, how low is the lens f-number (the lower the better) and (2) ease of installation and use. When covert operation is required in locations having widely varying light-level conditions or in a low light level application, a high sensitivity CCD solid-state or other intensified LLL camera used with a pinhole lens with an automatic iris controlling the light reaching the camera sensor is necessary. Shuttered CCD cameras may tolerate the use of manual-iris lenses. Check with the

manufacturer for the light range over which the camera will operate. Figure 18-5 shows straight and right-angle pinhole lenses with manual and automatic irises capable of controlling the light level reaching the camera sensor over a 35,000-to-1 light-level range.
A generic characteristic of almost all pinhole-type lenses is that they invert the video picture and therefore the cam-era must be inverted to get a normal right-side-up picture. Some right-angle pinhole lenses reverse the image right to left and therefore require an electronic scan-reversal unit (Section 16.4) to regain the correct left-to-right ori-entation. Some pinhole lenses have a focusing ring or the front element of the lens can be adjusted to focus a sharp image on the camera sensor.

18.3.2 Convertible Pinhole Lens Kit

Pinhole lenses have been manufactured for many years in a variety of focal lengths (3.8, 4, 5.5, 6, 8, 9, 11 mm), in straight, right-angle, and manual- and automatic-iris con-figurations. The FL of most of these lenses can be dou-bled to obtain one-half the FOV by using a 2X extender. Pinhole lenses with 16 mm and 22 mm FLs are achieved by locating a 2X magnifier in between the 8 and 11 mm lenses and the camera. This automatically doubles the f-number of each lens (only one-fourth of the light transmitted). In many applications, the required FLs and configuration are not known in advance, and the user (or dealer) must have a large assortment of pinhole lenses, or take the risk that he will not have the right lens to do the job. This dilemma was solved with the pinhole lens kit (Figure 18-6).

Eight different FL lenses can be assembled in either a straight or right-angle configuration within minutes with

· RIGHT-ANGLE SPRINKLER LENS ASSEMBLED FROM KIT

(B) LENS KIT IN CASE

FIGURE 18-6 Pinhole lens kit

this kit of pinhole lens parts. An additional four combina-tions can be assembled in the form of a disguised sprinkler-head covert application (Section 18.3.5). All lenses have a manual iris with automatic iris optional). Table 18-2 lists all the lens combinations for this versatile pinhole lens kit.

Covert Video Surveillance

451

Tables 18-3 and 18-4 tabulate the scene areas (width and height) as viewed with the popular pinhole lenses on 1/4 – and 1/3-inch sensor format cameras.

Several points should be considered when using stan-dard, fully assembled pinhole lenses or pinhole lenses made from the pinhole lens kit:

· Straight pinhole lenses invert the picture; therefore, the camera should be mounted in an inverted orientation.

· Some right-angle pinhole lenses will show a right-to-left picture orientation instead of left-to-right, as with normal lenses. A camera SRU will correct the problem. Check with the manufacturer.

· The straight pinhole lens with the sprinkler-mirror attachment displays a right-to-left picture. Use an elec-tronic SRU to correct the problem. The right-angle sprinkler-mirror version displays a correct left-to-right picture.

As an example: choose a pinhole lens and camera to view a scene 6 feet high by 8 feet wide at a distance of 15 feet using a 1/4 -inch format camera. Use Table 18-3 and choose an 11 mm FL lens. As another example, the scene area displayed on the monitor with an 8 mm lens on a 1/3-inch format camera in a ceiling at a distance of 20 feet is an area 22 feet wide by 16.4 feet high (Table 18-4).

Note that the FOV when using any of the medium- to long-FL lenses is independent of the hole size through which the lens views, providing the hole produces no tun-neling. Viewing through a wall with a wide-angle 4 –8 mm FL pinhole lens may require a cone-shaped hole or an array of small holes to prevent tunneling (vignetting) of the scene image.

18.3.3 Mini-Lenses

Mini-lenses and a mini-lens camera kit consisting of five interchangeable mini-lenses and a very small CCD camera are described in this section. Mini-lenses are small FFL objective lenses used for covert surveillance when space is at a premium (Figure 18.7).

FOCAL LENGTH (mm)

f/#

CONFIGURATION

IMAGE ORIENTATION

COMMENTS

11

2.3

STRAIGHT

NORMAL

PINHOLE LENS

8

2.0

STRAIGHT

NORMAL

PINHOLE LENS

11

2.5

RIGHT ANGLE

REVERSED

PINHOLE LENS

8

2.2

RIGHT ANGLE

REVERSED

PINHOLE LENS

22

4.6

STRAIGHT

NORMAL

PINHOLE LENS

16

4.0

STRAIGHT

NORMAL

PINHOLE LENS

22

5.0

RIGHT ANGLE

REVERSED

PINHOLE LENS

16

4.4

RIGHT ANGLE

REVERSED

PINHOLE LENS

11

2.3

STRAIGHT

NORMAL

SPRINKLER HEAD

22

4.6

STRAIGHT

NORMAL

SPRINKLER HEAD

11

2.5

RIGHT ANGLE

REVERSED

SPRINKLER HEAD

22

5.0

RIGHT ANGLE

REVERSED

SPRINKLER HEAD

Table 18-2 Pinhole Lens Kit Combinations and Parameters

452 CCTV Surveillance

1/4 inch SENSOR FORMAT LENS GUIDE

PINHOLE

CAMERA TO SCENE DISTANCE (D) IN FEET

LENS

WIDTH AND HEIGHT OF AREA (W ×H ) IN FEET

FOCAL

5

10

15

20

25

30

LENGTH

(mm)

W × H

W × H

W × H

W × H

W × H

W × H

2.6

12.3

× 9.2

24.6 × 18.5

36.9 × 27.7

49.2

× 36.9

61.5 × 46.1

74.0 × 55.5

3.7

8.5

× 6.5

17.3 × 13.0

30.0 × 19.5

34.6

× 26.0

43.3 × 32.5

60.0 × 39.0

4.0

8.0

× 6.0

16.0 × 12.0

24.0 × 18.0

32.0

× 24.0

40.0 × 30.0

48.0 × 36.0

6.2

5.2

× 3.9

10.4 × 7.8

15.6 × 11.7

20.8

× 15.6

26.0 × 19.5

31.2 × 23.4

8.0

4.0

× 3.0

8.0

× 6.0

12.0 × 9.0

16.0

× 12.0

20.0 × 15.0

24.0 × 18.0

9.0

3.6

× 2.7

7.2

× 5.4

10.8 × 8.1

14.4

× 10.8

18.0 × 13.5

21.6 × 16.2

11.0

2.9

× 2.2

5.8

× 4.4

8.7 × 6.6

11.6 × 8.8

14.5 × 11.0

17.4 × 13.2

16.0

2.0

× 1.5

4.0

× 3.0

6.0 × 4.5

8.0

× 6.0

10.0 × 7.5

12.0 × 9.0

22.0

1.5

× 0.8

2.9

× 2.2

4.4 × 3.3

5.8 × 4.4

7.3 × 5.5

8.7 × 6.6

Table 18-3 Pinhole Lens Guide for 1/4-Inch Format Camera

1/3-inch SENSOR FORMAT LENS GUIDE

PINHOLE

CAMERA TO SCENE DISTANCE (D) IN FEET

LENS

WIDTH AND HEIGHT OF AREA (W × H) IN FEET

FOCAL

5

10

15

20

25

30

LENGTH

(mm)

W × H

W × H

W × H

W × H

W × H

W × H

2.6

8.6 × 6.5

16.9 × 12.6

25.8 × 19.4

33.8 × 25.2

43.1 × 32.3

50.8 × 37.8

3.7

6.1 × 4.5

11.9

× 8.7

18.2 × 13.6

23.8 × 17.7

30.3 × 22.7

35.7 × 26.6

4.0

5.6 × 4.2

11.2

× 8.2

16.8 × 12.6

22.0 × 16.4

28.0 × 21.0

33.0 × 24.6

6.0

3.7 × 2.8

7.3

× 5.5

11.2 × 8.4

14.7 × 10.9

18.7 × 14.0

22.0 × 16.4

8.0

2.8 × 2.1

5.5

× 4.1

8.4 × 6.3

11.0 × 8.2

14.0 × 10.5

16.5 × 12.3

9.0

2.5 × 1.9

4.9

× 3.7

7.5 × 5.7

9.8 × 7.4

12.5 × 9.5

14.7 × 11.1

11.0

2.0 × 1.5

4.0

× 3.0

6.0 × 4.5

8.0 × 6.0

10.0 × 7.5

12.0 × 9.0

16.0

1.4 × 1.1

2.8

× 2.1

4.2 × 3.3

5.6 × 4.2

7.0 × 5.5

8.4 × 6.3

22.0

1.0 × .8

2.0

× 1.5

3.0 × 2.4

4.0 × 3.0

5.0 × 4.0

6.0 × 4.5

Table 18-4 Pinhole Lens Guide for 1/3-Inch Format Camera

Mini-lenses shown have focal lengths of 3.8, 8, 11 mm, etc. They have front-barrel diameters between 3/8 and 1/2 inch, making them easy to mount behind a barrier or in close quarters. Because these small lenses have no iris, they should be used in applications where the scene light level does not vary widely, or with electronically shuttered cam-eras. Mini-lenses, like other FFL lenses and unlike pinhole lenses, do not invert the image on the camera. Since the small and short (less than 5/8 inch long) mini-lenses have only three to six optical lens elements, fast optical speeds of f/1.4 to f/1.8 are realized. Pinhole lenses, on the other hand, are 3–5 inches long, and have as many as 10–20 optical elements and optical speeds of f/2.0 to f/4.0. This makes the mini-lens approximately five times faster (able to collect five times more light) than the pinhole lens.

18.3.3.1 Off-Axis Optics

A useful variation of the mini-lens is one that is mounted with its optical axis laterally offset from the camera-sensor axis (Figure 18-8). This offset configuration allows the camera to view a scene at an angle away from the camera-pointing axis. The physical amount the optics must be moved to produce a large offset angle is only a few millime-ters, which is easily accomplished with this special mini-lens and its modified mount. The offset angle is chosen so that, with the camera parallel to a mounting surface, the entire lens FOV views the scene of interest without view-ing the mounting surface. This angle is 22 for the 8 mm lens and 15 for the 11-mm when using a 1/4 -inch format camera. It is 18 and 13 , respectively for the same lenses when using a 1/3-inch camera. This technique has a direct

Covert Video Surveillance

453

CAMERA

1/3″ FORMAT CAMERA

SCENE
MINI-LENS

8 mm
lens on camera

11 mm FL

3.8 mm FL

FIGURE 18-7 Mini-lens and optical diagram

ON AXIS

CAMERA

MINI-LENS

LENS OFFSET

FOV

ON-AXIS

LENS

OFF-AXIS

MINI-LENS

CCD SENSOR

FOV

OFF-AXIS

LENS

CEILING MOUNTED CAMERA

CEILING

SIDE VIEW

WALL MOUNTED CAMERA

8 mm OFF-AXIS

WALL

MINI-LENS

TOP VIEW

FIGURE 18-8 Off-axis optics configuration

454 CCTV Surveillance

benefit when a camera/lens is mounted flat against a wall or a ceiling or other mounting surface (Figure 18-8).

18.3.3.2 Optical Attenuation Techniques

Since mini-lenses do not have an iris, they should be used when the lighting conditions are fairly constant and do not exceed the dynamic range of the camera. If the scene is very brightly illuminated with an intense artificial light or the sun, several techniques can be used to attenuate the light to the lens/camera (Figure 18-9).

The first technique is to mount the mini-lens behind a light-attenuating filter (Figure 18-9a). This may take the form of a gray neutral-density filter, a partially alu-minized film, or a tinted/smoked glass or plastic material. Neutral-density filters are available from photographic sup-ply stores. This technique uniformly attenuates the light across the full aperture of the lens. A second technique shown in Figure 18-9b through 18-9e is to mount the mini-lens behind a small hole, a pattern of small holes, a slit, or other hole(s). This is accomplished by either mount-ing a small cap with the hole(s) (Figure 18-9b) onto the lens, or mounting the lens behind a hole(s) in the barrier (Figure 18-9c –18-9e). The light level reaching the camera sensor can be set initially by locating the lens behind a hole smaller than the mini-lens diameter. This technique

attenuates the light reaching the lens but does not do it uniformly. For medium-FL lenses (11 mm and above), almost any shape hole results in a satisfactory image on the sensor. When the 11 or 22 mm mini-lens or pinhole lens is mounted behind a viewing barrier, a central hole as small as 1/16th of an inch is suitable for producing a full image of the scene, providing sufficient light avail-able for the camera. When short focal length (2.2, 3, 8 mm, etc.) mini-lens or pinhole lens views through a small hole, an undesirable porthole effect occurs, which is eliminated by having the lens view through a central hole and a series of concentric holes located around the central hole. The hole pattern must extend to the outer limits of the lens so that the full FOV of the lens is maintained. These concentric holes enable the lens to have peripheral vision or wide-angle viewing, and they eliminate vignetting. Figure 18-9(b, d) shows two examples of this extended hole pattern. Either technique can provide attenuations required for sunlit or brightly illuminated scenes.

18.3.3.3 Mini-Camera/Mini-Lens Combination

A high-sensitivity pinhole camera results when a very fast mini-lens—f/1.4 to f/2.0—is coupled directly with the camera sensor. Figure 18-10 illustrates a mini-lens camera kit with three standard on-axis mini-lenses having focal

(A) UNIFORM LIGHT ATTENUATION

ACROSS LENS APERTURE: (C)

• NEUTRAL DENSITY FILTER
• SMOKED OR TINTED GLASS

OR PLASTIC

(D)

(B–E) DISCRETE APERTURE ATTENUATOR

· SINGLE HOLE

· MULTIPLE HOLES

• SLIT(S) (E)

(B)

FIGURE 18-9 Lens optical attenuation techniques

6 MINI-CAMERA MINI-LENS 3.8 mm FL ASSEMBLED FROM KIT

12 MINI-CAMERA MINI-LENS KIT

FIGURE 18-10 Mini-camera/mini-lens combination and kit

lengths of 3.8, 8 and 11 mm and two off-axis mounts for the 8 and 11 mm FL lenses, and a very small, sensitive, high-resolution color CCD camera. The complete camera is only 125 × 125 × 100 inches long. The 11-mm FL lens extends 0.3 inch in front of the camera. The camera oper-ates directly from 12 volts DC, requires only 1.5 watts of power, and produces a standard composite video output.

The small lens size and direct coupling to the camera sensor do not leave room for a manual or automatic iris. The camera has excellent electronic light-level compensa-tion, but optimum performance is achieved if the lighting is fairly constant. Under bright light conditions an atten-uation technique shown in Figure 18-9 is used.

Covert Video Surveillance

455

18.3.4 Comparison of Pinhole Lens and Mini-Lens

To compare different pinhole and mini-lenses with respect to their ability to transmit light to the camera sensor, a light power factor (LPF) is defined, with a slow pinhole lens (f/4.0) as a base reference. Table 18-5 summarizes the optical speed (f-number) and LPF for standard pinhole and mini-lenses.

The f-number is usually critical in nighttime applications with low light levels and where auxiliary lighting cannot be added. Table 18-5 illustrates the significantly higher amount of light passing through the mini-lenses as com-pared with the pinhole lenses. A camera/lens using an f/1.8 mini-lens transmits almost five times as much light to the camera sensor as an f/4 pinhole lens. The f/1.4 mini-lens transmits more than eight times as much light as the f/4 pinhole lens.

18.3.5 Sprinkler-Head Pinhole Lenses

A very effective covert system uses a camera and lens camouflaged in a ceiling-mounted sprinkler head. Of the large variety of covert lenses available for the security video industry (pinhole, mini, fiber-optic), this unique, extremely useful product hides the pinhole lens in a ceiling sprinkler fixture, making it very difficult for an observer standing at floor level to detect or identify the lens and camera. Figure 18-11a shows the sprinkler pin-hole lens attached to a standard camera mounted on a ceiling.

The covert surveillance sprinkler installed in the ceil-ing in no way affects the operation of the active fire-suppression sprinkler system; however, it should not be installed in locations that have no sprinkler system, so as not to give a false impression to fire and safety personnel that there is a sprinkler system installed.

The only part of the lens system visible from below is the standard sprinkler head and the small (3/8×5/8-inch) mirror assembly. In operation, light from the scene reflect-ing off the small mirror is directed by the mirror to the front of the pinhole lens. The 11 or 22 mm pinhole lens transmits and focuses the scene onto the camera sensor. In the straight version the image is reversed. In surveil-lance applications this is often only an annoyance and not really a problem. However, if it needs to be corrected an electronic SRU will correct this condition. The right-angle version (Figure 18-11b) corrects this condition and pro-duces a normal left-to-right image scan. The small mirror can be adjusted in elevation to point at different scene heights. To point in a particular azimuth direction, the entire camera-sprinkler lens assembly is rotated with the mirror pointing in the direction of the target of interest. When installed, most of the pinhole lens and the entire

456

CCTV Surveillance

FOCAL

LIGHT

ANGULAR FOV(°)

POWER

COMMENTS

LENGTH

f/#

LENS TYPE

CONFIGURATION

1/4-inch FORMAT

1/3-inch FORMAT

FACTOR

(mm)

(LPF)

*

HORIZ

VERT

HORIZ

VERT

2.6

2.0

MINI

STRAIGHT

4.0

62.4

46.8

83.2

62.4

ULTRA WIDE-ANGLE

3.8

1.4

MINI

STRAIGHT

8.16

39.1

31.8

52.1

42.4

ULTRA WIDE-ANGLE

8.0

1.6

MINI

STRAIGHT

6.25

21.8

16.7

29.1

22.3

LONG TAPER

11.0

1.8

MINI

STRAIGHT

4.94

16.2

12.3

21.6

16.4

LONG TAPER

25.0

4.0

MINI

STRAIGHT

1.0

7.0

5.3

9.3

7.1

ULTRA WIDE-ANGLE

3.8

2.0

PINHOLE

STRAIGHT

4.0

39.1

31.8

52.1

42.4

ULTRA WIDE-ANGLE

3.8

2.2

PINHOLE

RIGHT-ANGLE

3.31

39.1

31.8

52.1

42.4

ULTRA WIDE-ANGLE

5.5

3.0

PINHOLE

STRAIGHT

1.78

32.3

25.6

43.1

34.1

WIDE-ANGLE

6.2

2.0

PINHOLE

STRAIGHT

4.00

28.0

21.5

37.3

28.7

WIDE-ANGLE

8.0

2.0

PINHOLE

STRAIGHT

4.00

21.8

17.7

29.1

23.6

SHORT, WIDE-ANGLE

8.0

2.2

PINHOLE

RIGHT-ANGLE

3.31

21.8

17.7

29.1

23.6

LONG TAPER

9.0

3.5

PINHOLE

STRAIGHT

1.31

19.4

15.1

25.9

20.1

LONG TAPER

11.0

2.3

PINHOLE

STRAIGHT

3.02

16.2

12.3

21.6

16.4

SHORT TAPER

11.0

2.5

PINHOLE

RIGHT-ANGLE

2.56

16.2

12.3

21.6

16.4

SHORT TAPER

16.0

4.0

PINHOLE

STRAIGHT

1.00

11.0

8.3

14.7

11.1

NARROW-ANGLE

11 INCREASE IN LIGHT LEVEL REACHING SENSOR BASED ON USING VALUE OF 1.00 FOR AN f/4 PINHOLE LENS

Table 18-5 Pinhole Lens and Mini-Lens Light Transmission Comparison

camera is concealed above the ceiling, with only a modi-fied sprinkler head, a small mirror, and small lens in view. For many applications this stationary pinhole lens pointing in one specific direction is adequate. To look in different directions the camera, sprinkler head, and moving mir-ror assembly are made to pan (scan) via a motor drive. A motor drive sprinkler scanning system can provide remote panning capability. A scanning version of the sprinkler concept has a remote-control 360 continuous panning capability (Figure 18-12).

18.3.6 Mirror-Pinhole Lens

Large plastic domes are often used to conceal a PTZ video surveillance system from the observer (Chapter 14). The purpose for concealing the camera and lens in the dome is so that the observer cannot see the direction in which the camera lens is pointing or whether there is actually a surveillance camera. Using this subterfuge, one camera system can scan and view a large area without the observer knowing at any instant whether he is under observation. Most domes are from 5 to 10 inches in diameter and drop below the ceiling by 5–8 inches. The requirement that the lens view through the dome results in a typical light loss of 50%. A more covert camera/lens assembly takes the form of a camera, pinhole lens, and small mirror.

If the right angle lens of the sprinkler-head assembly shown in Figure 18-11 and 18-12 is removed, all that pro-trudes below the ceiling is a small mirror approximately

3/8 × 5/8 inches. This technique results in a very low pro-file that is difficult for an observer to detect at ground level. The pinhole/mirror system provides an alternative to some dome applications. The system can be fixed or have a 360 panning range.

Two advantages of the moving mirror system over the dome are: (1) no large protruding dome suspended below the ceiling and (2) easy installation. Installation is easy since only a small hole about 3/4 inch in diameter is required to insert the lens and mirror through the ceil-ing. The small mirror scanning system has limitations: (1) it cannot view the scene directly below its location and

27 there is no zooming. The dome system has two advan-tages over the scanning mirror: (1) the dome serves as a deterrent since the observer sees the dome and believes a camera is active in it but does not know at any instant where the camera is looking, and (2) the added capability of full-range zoom optics.

18.3.7 Fiber-Optic Lenses

When the barrier between the scene side and the camera/ lens side is a few inches as in Figure 18-3, a pinhole or mini-lens and camera can be mounted directly behind the bar-rier. For difficult covert video surveillance applications in which small cameras and mini pinhole lenses will not work, coherent fiber-optic-bundle lenses may be the solution.

Fiber optics are used when it is necessary to view a scene on the other side of a thick barrier or inside a confined area. The fiber-optic bundle lens and camera are installed

Covert Video Surveillance

457

(A) STRAIGHT

(B) RIGHT-ANGLE

FIGURE 18-11 Sprinkler-head pinhole lenses

behind the barrier and the objective lens on the scene side. The lens viewing the scene can be a few inches or a few feet away from the camera. There are three opti-cal techniques to transfer the image, in effect “lengthen” the camera’s objective lens: (1) a rigid coherent fiber-optic conduit, (2) a borescope lens, and (3) a flexible fiber-optic bundle. These special lenses can extend the objective lens several inches to several feet in front of the camera sensor. The rigid fiber conduit uses a fused array of fibers and cannot be bent. The flexible fiber lens has hair-like fibers loosely contained in a protective sheath and can be flexed and bent easily. These fiber-optic lenses should not be confused with the single or multiple strands of fiber commonly used to transmit the time-modulated video signal a long distance from a camera to a remote monitoring site (Chapter 6). Coherent fiber-optic lenses typically have 200,000–300,000 individual fibers forming an image-transferring array. Rigid fiber-optic lenses are 1/4 –1/2 inch in diameter and from 6 to 12 inches long. Flexible fiber-optic lenses are from 1/8 to 1¼ inch in diam-eter and up to several feet long. These fiber-optic lenses are available with manual or automatic iris for 1/6-, 1/4-, 1/3-, 1/2-, and 2/3-inch video formats.

By combining lenses with coherent fiber-optic bun-dles, a long, small-diameter optical lens is produced that requires a small hole for insertion into the barrier. A small aperture hole is drilled completely through at the barrier

surface and connected to the camera on the protected side (Figure 18-13).

This lens/camera system has provided the solution for many banking ATM and correctional-facility security prob-lems. A minor disadvantage of all fiber-optic systems is that the picture obtained is not as “clean” as that obtained with an “all-lens” pinhole lens. These imperfections occur because several hundred thousand individual hair-like fibers make up the fiber-optic bundle some of which are not perfectly transmitting. For most surveillance applica-tions the imperfections do not result in any significant loss of intelligence in the picture. Figure 18-14 shows complete rigid and flexible fiber-optic lenses.

18.3.7.1 Configuration

A fiber-optic lens consists of three parts: (1) an objec-tive lens that focuses the scene onto the front end of the fiber-optic bundle, (2) a rigid conduit or flexible fiber coherent optic bundle that transfers the image a substan-tial distance (several inches to several feet), and (3) a relay lens at the output end of the fiber bundle that re-images the output image and focuses onto the camera sensor (Figure 18-15).

The objective lens can be like any of the FFL, zoom, pinhole, manual-, or automatic-iris lens. The objective lens

458 CCTV Surveillance

PANNING MECHANISM

ABOVE CEILING

CAMERA

RIGHT

ANGLE

CAMERA

PINHOLE

PINHOLE

LENS

LENS

ADJUSTABLE

MIRROR

270°

360°

PANNING

CONTINUOUS

PANNING

FIGURE 18-12 Panning sprinkler-head pinhole lens system

SCENE

CAMERA

3/16″ TO 1/2″

DIAMETER HOLE

RIGID FIBER OPTIC

MANUAL

IRIS

(6–12 INCHES LONG)

THICK WALL BARRIER

(6–12 INCHES)

FIGURE 18-13 Fiber-optic pinhole lens installation in thick wall

Covert Video Surveillance

459

OBJECTIVE LENS: 8 mm OR 11 mm FL

FIBER TYPE: RIGID CONDUIT

FIBER LENGTH: 6 inches

RELAY LENS: M = 1:1

IRIS: MANUAL

MOUNT: C OR CS

(A) RIGID CONDUIT LENS

OBJECTIVE LENS: ANY C OR CS MOUNT

FIBER TYPE: FLEXIBLE BUNDLE

FIBER LENGTH: 39 inches

RELAY LENS: M = 1:1

IRIS: MANUAL

MOUNT: C OR CS

(B) FLEXIBLE BUNDLE LENS

FIGURE 18-14 Rigid and flexible fiber-optic lenses

must produce an image large enough to fill the full aper-ture (cross-sectional area) of the fiber-optic bundle. The coherent fiber-optic bundle consists of several hundred thousand closely packed glass fibers to coherently transfer an image from one end of the fiber to the other, several inches to several feet (Figure 18-16).
Fiber 1 transmits point 1 of the image from the objec-tive lens down the fiber to a corresponding point 1 on the exit end of the fiber bundle. Likewise, all of the remain-ing points of the entrance image are transferred in an exact one-to-one correspondence to the exit end of the fiber bundle, thereby producing a coherent image. Coher-ent means that each point in the image on the front end of the fiber bundle corresponds to a specific point at the rear end of the fiber bundle.

18.3.7.2 Rigid Fiber Pinhole Lens

The rigid fiber-optic bundle has individual fibers that are fused together to form a rigid glass rod or conduit and is

usually protected from the environment and mechanical damage by a rigid metal tube (Figure 18-14). The fiber-optic bundle is approximately 0.4 inch in diameter for a 2/3-inch format sensor, 0.3 inch for a 1/2 inch, 0.2 inch for a 1/3 inch, and 0.15 inch for a 1/4 inch. For the 2/3 inch format, the outside diameter is about 0.5 inch. It should be noted that the image exiting the fiber-optic lens is inverted with respect to the image produced by a standard objective lens. This inversion is corrected by inverting the camera. The fiber-optic lens speed is between f/4 and f/8 depending on the fiber length—slow in comparison with the standard, all-lens type pinhole lens.

18.3.7.3 Flexible Fiber

When the most flexibility between the front objective lens and the camera is required, an alternative to the remote-head CCD camera is a coherent flexible fiber-optic bundle (Figure 18-14). The front of the flexible fiber-optic bun-dle has a C mount and accepts any pinhole, C, or CS

460 CCTV Surveillance

RELAY LENS

OBJECTIVE LENS

CAMERA

SCENE

COHERENT

SENSOR

FIBER OPTIC BUNDLE

6 –12 inches LONG

IMAGE

ON

SENSOR

FIBER

SCENE

OPTIC

IMAGE

OUTPUT

SCENE

FIGURE 18-15 Fiber-optic lens configuration

FLEXIBLE FIBER
BUNDLE ENDS
EPOXIED
RIGID
CONDUIT FIBER #1
12 MICRON FIBERS
LOOSELY HELD IN
PROTECTIVE SHEATH

FIBER #1

FIBER #1

FUSED

RIGID
GLASS
FIBERS

FIBER
ENDS
EPOXIED

FIBER #1

FIGURE 18-16 Fiber bundle construction

mount lens. The rear lens terminates in a male C mount, suitable for any C or CS mount camera. One advantage the fiber-optic lens has over a remote head camera is that there is no electrical connection from the front objective lens to the camera sensor, which may be important in some applications, for example environmental protection (from adverse weather, corrosive environment, or mechan-ical abuse). It can be twisted through 360 with no image degradation. It, too, has spots like the rigid fiber-optic. The flexible fiber-optic lens has a 180 “twist” built into it and therefore does not invert the picture. The flexible fiber-optic bundle individual fibers are fused together only at the ends, but are free to move in the length between the ends.

18.3.7.4 Image Quality

As shown in Figure 18-16, the fiber-optic bundle is assem-bled from several hundred thousand individual glass fiber-optic strands. Although high technology and careful assembly techniques are used throughout the fiber bundle manufacturing process to achieve maximum uniform opti-cal transmission, there are small variations in transmission from one fiber to another and some broken fibers. The result is that in almost all fiber-optic systems, the picture obtained is not as “clean” as that obtained with an “all-lens” pinhole lens. There are some cosmetic imperfections that look like dust spots (actually non- or partially transmitting fibers), as well as a geometric pattern caused by pack-ing the fibers during manufacture. These imperfections occur because there are several hundred thousand indi-vidual hair-like fibers comprising the fiber-optic bundle, and some of them are not transmitting perfectly. For many

Covert Video Surveillance

461

applications these imperfections do not result in any loss of picture intelligence, making the lens system adequate for identification of people, actions, and other informa-tion. Some fiber-optic lenses have a resolution of 450–500 TV lines, similar to a high-quality 1/4-, 1/3-, and 1/2-inch camera system. Figure 18-17 shows two examples of images produced from a rigid and flexible fiber-optic lens.

The photographs were taken directly from a 9-inch monochrome monitor using a CCD solid-state camera with resolution of 570 horizontal TV lines. Figure 18-17a shows the typical resolution and image quality obtainable from a 1-meter, flexible fiber-optic lens: approximately 450 TV lines horizontal and 350 vertical. The spots are caused by partially transmitting or non-transmitting fibers. Figure 18-17b shows the same image obtained with an 8-inch rigid fiber-optic lens. The vignetting at the corners of the image was caused by the relay lens, not the fiber bundle. Note the spots and honeycomb pattern in the rigid fiber-optic monitor picture. The honeycomb is caused by the fiber-stacking procedure and consequent heat fusing of the rigid bundle.

18.3.8 Bore-Scope Lenses

The bore-scope lens viewing system is a long thin tube housing with multiple relay lenses used to view inside objects (such as safes) or through barriers. Bore-scope sizes range from 12 to 30 inches long, and from 1/8 to 3/8 inch in diameter (Figure 18-18).

Special mini-bore-scopes are available with 1–2 mm outside diameters, 2–6 inches long. Bore-scopes are constructed from stainless-steel tubing and contain an

(3) FIBER: FLEXIBLE: 39 inches LONG

OBJECTIVE LENS: 25 mm FL, F/1.4

RELAY LENS: M = 1:1

OVERALL F/#: 4.0

· FIBER: RIGID: 6 inches LONG

OBJECTIVE LENS: 8 mm FL, F/1.6

RELAY LENS: M = 1:1

OVERALL F/#: 6.0

FIGURE 18-17 Resolution and image quality from fiber-optic lenses

462 CCTV Surveillance

FIGURE 18-18 Boroscope lens viewing system

CAMERA

CAMERA

RELAY

OPTICS

TOTAL

PROBE TUBE

SCAN

20°

DIAMETER: 9/16″

106°

SCAN MIRROR

WORKING LENGTH: 18″–50″

“all-lens” optical system. The long lengths and all-lens design mandate that such lenses have very high f-numbers: they are optically slow. Typical designs have an f-number between f/15 and f/40. By comparison, an f/5 lens trans-mits 16 times more light than an f/20 lens. The bore-scope must be used with high levels of lighting or an LLL camera (Chapter 19).

18.4 SPECIAL COVERT CAMERAS

18.4.1 PC-Board Cameras

The miniaturization of 1/6 -, 1/4 -, 1/3 -, and 1/2 -inch CCD and complimentary metal oxide semiconductor (CMOS) sensors and camera electronics has generated a new family of small single and dual printed-circuit (PC) board surveillance cameras. Three PC-board and housed flat cameras are shown in Figure 18-19.

Figure 18-19a shows a color camera with a CS mount and automatic-iris option. Figure 18-19b shows a 1/3-inch format PC-board CCD camera with an 8 mm FL mini-lens and six IR LEDs for night-time illumination. Other inter-changeable lenses—3.8, 5.5, and 11 mm FL—are avail-able. Figure 18-19c shows a compact flat camera sealed in a metal/epoxy case with pin terminals at the rear. The 1/3-inch format camera has 380-TV-line resolution and 0.2-fc sensitivity. All cameras are powered by 12 volts DC.

18.4.2 Remote-Head Cameras

The small size of mini-lenses and CCD and CMOS camera sensors permits the construction of extremely small covert

lens-sensor heads by remoting the lens and sensor from the camera electronics via a small electrical cable. The cable link between the camera head and the camera electronics can vary from a few inches to 100 feet. Figure 18-20a shows a monochrome 1/3-inch format CCD remote-head cam-era with an 11-mm FL, f/1.8 lens, and an 18-inch cable connecting the sensor-lens with the camera electronics.

The camera has a resolution of 450 TV lines and a light sensitivity of 0.1 fc. Figure 18-20b shows a small color CCD remote-head camera with a 7.5 mm FL, f/1.6 lens on a 1/2-inch format sensor. The lens-sensor head is 0.69 inch in diameter × 225inches long and weighs only 0.64 ounce. The camera has a resolution of 460 TV lines and a sensitivity of 1.0 fc.

18.5 INFRARED COVERT LIGHTING

Video surveillance augmented with invisible IR covert lighting can significantly increase the usefulness of covert installations. Since the covert camera is intended to be hid-den from its target, if the covert video system can operate in near or total darkness the person under surveillance will not be aware that he is under observation. By augmenting the camera system with an IR light, invisible to the human eye but not to the camera, the resulting video image can be as good as that obtained under normal visible daylight conditions. CCD, CMOS, and other LLL cameras are sensi-tive to this IR radiation and can “see” with this IR lighting. The amount of IR radiation the camera responds to and the resulting quality of the picture depends on the type of IR lamp or LED used, its power level and beam angle (Chapter 3), and the sensitivity of the camera to the IR radiation. This last factor depends on whether an IR cut

Covert Video Surveillance

463

(A) 1/3″ COLOR CMOS (B) 1/4″ COLOR CCD

(C) HIGH RESOLUTION 1/2″ COLOR CCD

FIGURE 18-19 Flat printed circuit PC-board cameras

filter is in place in the camera and on the CCD sensitivity to the IR energy.

18.5.1 Concealment Means

Light sources that emit both visible and IR light (tungsten, tungsten-halogen, xenon lamps, and others) can be opti-cally filtered so that only the IR radiation leaves the source and irradiates the scene. High-efficiency, low-power LED semiconductors produce sufficient IR energy to illuminate an area suitable for covert operation while being invisible to the eye. Figure 18-22 illustrates the principle and several techniques of producing IR illumination.

The thermal lamp or LED source emits IR radiation that reflects off the scene and off objects in it. The lens and camera collect the reflected IR energy to produce a video image signal. The IR-emitting source is often con-cealed by installing it behind an opaque (tinted) plas-tic or one-way (partially aluminized) window. Another technique is to use a spectral beam-splitting window that

transmits the invisible IR radiation and blocks the visible radiation. Another technique is to conceal the IR-emitting source just as the pinhole lens is concealed, by locating the source at the focal plane of a pinhole lens and directing the energy at the same target the pinhole lens is viewing. Usually the beam from the pinhole lens IR source is made slightly larger than the FOV of the pinhole lens–camera combination. Alignment is necessary between the camera and IR source since the IR beam must illuminate the same scene the pinhole lens is looking at. When the application is to perform covert surveillance at short distances and in small rooms (10–15 feet), a wide-area IR illuminator is used since the alignment is not critical.

18.5.2 IR Sources

There are numerous commercially available thermal lamp and LED IR sources for covert surveillance applica-tions. They vary from short-range, low-power, wide-angle beams to long-range, high-power, narrow-angle beam

464 CCTV Surveillance

(A) ULTRA SMALL 1/4″ DIA.

(B) HIGH RESOLUTION

FIGURE 18-20 Remote head cameras

types. Figure 18-23 illustrates two IR LED and thermal IR source illuminators.

A single IR LED emits enough IR energy to produce a useful picture at ranges up to a few feet with a CCD camera. By stacking many (10 to several 100) LEDs in an array, higher IR power is directed toward the scene, and a larger area at distances up to 50–100 feet may be viewed (Figure 18-23a). Filtered thermal lamp IR sources with power levels up to several hundred watts can illumi-nate large areas at distances up to several hundred feet (Figure 18-23b). These are usually used in outdoor appli-cations where longer ranges are required and personnel cannot come into close proximity to them. Since the radi-ation source is not visible to the human eye personnel should not come in close proximity to them.

18.6 LOW-LIGHT-LEVEL CAMERAS

The camera parameter most critical to the successful view-ing of a scene under low light level (LLL) conditions with a covert system is the camera sensor sensitivity. Most monochrome CCD cameras have sensitivities of approxi-mately 0.2–1 fc (0.1 lux), which does not result in satisfac-tory CCTV picture quality under dawn, dusk, nighttime, or poorly lighted indoor conditions. A few special CCD cameras produce sensitivity of 0.003 fc (0.0003 lux) which

substantially increases its usefulness at low light levels. It also boasts a resolution of 570 TV lines.

When CCD camera sensitivity is not sufficient and addi-tional lighting cannot be added, a LLL camera such as an intensified CCD (ICCD) or intensified SIT (ISIT) must be used (Chapter 19). These light-intensified cameras oper-ate at significantly lower light levels than the solid-state cameras. The newer ICCD camera has a sensitivity match-ing that of the prior generation SIT camera. All this increased sensitivity comes at a cost. Any intensified cam-era is expensive and should be considered only for critical security applications.

18.7 IMBEDED COVERT CAMERA CONFIGURATIONS

Video cameras and lenses are concealed in many different objects and locations including overhead track lighting fixtures, emergency lighting fixtures, exit signs, tabletop radios, table lamps, wall or desk clocks, shoulder bags, and attaché cases (Figure 18-21).

Figure 18-21a shows a popular emergency light that was modified to house a camera and mini-lens system with the camera viewing from behind the front bezel. The emer-gency lighting fixture operates normally, can be tested for operation periodically, and its operation is in no way affected by the installation of the camera. The housing has an angled extension that points the housing downward by about 15 so that the lens points downward and optimally views the area. Alternatively an off-axis mini-lens could be used instead of the on-axis mini-lens to make the cam-era look downward. The lens views through the smoked (tinted) plastic front window and cannot be seen by an observer even at close range.

The exit light fixture is another convenient housing for camouflaging a covert camera system (Figure 18-21b). A wide-angle mini-lens on a small PC-board camera is all that is required for this covert camera installation.

A wall-mounted clock is an ideal location for camouflag-ing a covert camera/lens combination (Figure 18-21c). The lens views out through one of the black numerals. In this case, the flat camera (approximately 7/8 inch deep) and mini-lens are mounted directly behind the numeral 11 on the clock. The camera uses offset optics (Figure 18-8) so that the camera views downward at approximately a 15 angle even though the clock and camera are mounted vertically on the wall.

Figure 18-23d shows a no smoking sign into which a camera and lens have been installed. The camera views through an imperceptible hole in the sign. Figure 18-24 shows a ceiling-mounted sprinkler-head camera. An option to any of these covert cameras is a wireless RF or microwave transmitter. These covert camera systems can also be designed using a digital IP wireless camera and viewed using an Internet browser. The items into which

Covert Video Surveillance

465

(A) EMERGENCY LIGHT (B) CLOCK
CAMERA VIEWS THROUGH CAMERA VIEWS THROUGH
BLACK PLASTIC HOLE AT NUMERAL “11”

(C) EXIT SIGN (D) NO SMOKING SIGN
CAMERA VIEWS THROUGH B&W CAMERA VIEW THROUGH
HOLE IN EITHER ARROW BLACK OPAQUE PLASTIC

FIGURE 18-21 Covert cameras installed in office building fixtures

covert cameras can be installed are limited only by the imagination of the user.

18.8 WIRELESS TRANSMISSION

The video signal from the covert camera is sent to the monitor, VCR, DVR, or over the Internet via RG59/U 75-ohm coaxial cable, UTP, LAN, WAN, or wireless LAN (WiFi). If a dedicated telephone-grade line (two-wire) is available, the UTP using a special line driver and receiver pair provide good transmission of a real-time video signal over several thousand feet of continuous telephone wire (Chapter 6). For digital video transmission CAT-5e cable is used.

Covert video applications often require that the cam-era/lens system be installed and removed quickly, or that it remain installed on location for only short periods of time. This may mean that a wired transmission link (such as coaxial cable or fiber -optic) cannot be installed

and a wireless transmission link from camera to moni-tor or recorder is required. This takes the form of a low power radio frequency (RF) or microwave video transmit-ter mounted near the video camera. A description of these transmitters is given in Chapter 6, but those specifically applicable to covert applications are summarized here. The RF transmitters are less than 100 milliwatts output and transmit the video images over ranges from 100 to 2000 feet. In the United States, the FCC restricts the use of the higher-power transmitters to federal or government agencies and allows only low-power units for commercial or industrial use.

Figure 18-25a shows a low-power RF, 100-mw transmit-ter and receiver operating at 920 MHz that can transmit an excellent monochrome or color video picture over a distance of a few hundred feet.

Figure 18-25b shows a 2.4 GHz microwave transmitter that transmits excellent monochrome and color images over distances up to a few hundred feet indoors and 2000 feet outdoors. Using a directional (Yaggi) receiver antenna can increase the range further. While RF and

466 CCTV Surveillance

(A) LAMP WITH FILTER

TUNGSTEN LAMP

TUNGSTEN HALOGEN LAMP

SPOT OR FLOOD

IR TRANSMITTING

LAMP (PAR)

FILTER

METAL HOUSING

WITH COOLING

FINS (HEAT SINK)

AND CONVECTION

SWIVEL

MOUNT

(B) LED ARRAY

(C) CCTV LENS WITH IR FILTER

LED

LENS

IR

FILTER

IR

FFL OR

SOURCE

ZOOM

CCTV

LENS

IR

PINHOLE

IR

SOURCE

LENS

FILTER

FIGURE 18-22 IR illumination technique

(A) IR LED ARRAY (B) IR THERMAL

FIGURE 18-23 IR source illuminators: IR LED array, IR thermal lamp

FIGURE 18-24 Sprinkler-head covert camera

microwave transmitters can be used indoors, recognize that these frequencies cannot pass through metal objects and therefore the systems should be tested on site, through a steel building or near other metallic or reinforced con-crete structures before an installation is made. While the transmitter may have suitable range under outdoor, unobstructed conditions, when used indoors or between two points with obstructions, the only way to determine the useful range of the link is to put the system into operation. The deleterious effects most readily observed are: (1) reduction in range, (2) ghost images (multiple images produced by reflections of the signal from metallic objects), and (3) unsynchronized pictures (picture breaks up). Repositioning the transmitter or receiver equipment often substantially improves or eliminates such problems. Most microwave systems have a more directional transmit-ting pattern than RF transmitters. This means the antenna directs the energy toward the receiver, and therefore align-ment between transmitter and receiver is more critical. Most microwave installations are line of sight but the microwave energy can be reflected off objects in the path between the transmitter and the receiver to direct the energy to the receiver, at a sacrifice in range. The higher frequency of operation and directionality make microwave installation and alignment more critical than the RF trans-mitters (Chapter 6).

Commercial microwave transmission systems operate in the 2.4 and 5.8 GHz frequency range and do not require FCC licensing and approval. Other frequencies can only be used by government agencies and some commercial customers if they apply to the FCC for a license. One condition in obtaining approval is to have a frequency search performed to ensure that the system causes no interference to existing equipment in the area.
Another line-of-sight system requiring no FCC approval is a wireless gallium arsenide (GaAs) IR optical trans-mission system. This light-wave system requires no cable connection between the transmitter and the receiver and achieves ranges of hundreds to several thousands

Covert Video Surveillance

467

(A) RF TRANSMITTER

(B) MICROWAVE

FIGURE 18-25 RF and microwave transmitters for covert video transmission

of feet (Chapter 6). Its major limitation is the severe reduction in range under fog or heavy precipitation conditions.

18.9 COVERT CHECKLIST

· Optical speed or f-number is probably the most impor-tant reason for choosing one pinhole lens over another. The lower the f-number the better. An f/2 lens trans-mits four times more light than an f/4. This can mean the difference between using a standard CCD or CMOS camera and using a LLL ICCD.

468 CCTV Surveillance

· Most pinhole lenses have a FL between 3.8 mm and 22 mm and are designed for 1/4- and 1/3-inch format cameras. Tables 18-1, 18-3, 18-4, and 18-5 show the FOVs obtained with these lenses. For example, using these tables or the Lens Finder Kit (Chapter 4), the FOV seen

with the 11 mm lens on a 1/3-inch camera format at a distance of 15 feet is an area 6 feet wide by 4.5 feet high displayed on the monitor. Note that the FOV is independent of the hole size through which the lens views, providing a hole produces no tunneling. When viewing through a wall with a wide-angle pinhole lens or mini-lens (3.8, 5.5, or 8 mm), the lens may require a cone-shaped hole or an array of small holes to prevent tunneling (vignetting) of the scene image.

· A short FL lens (3.8 mm) has a wide FOV and low mag-nification. A long FL lens (25 mm) has a narrow FOV and has high magnification.

· Medium FL lenses produce FOVs wide enough to see much of the action and still have enough resolution to identify the persons or actions in the scene. A short FL lens sees a wide FOV and objects are not well resolved. Long FL lenses see a narrow FOV with objects well resolved (clear).

· Under most conditions, the small-barrel, slow-taper pin-hole lens is easier to install and is the preferred type over the wide-barrel, fast-taper shape. The user must weigh the pros and cons of both types.

· The use of a straight or right-angle pinhole lens depends on the space available behind the barrier for mounting the lens and camera, and on the pointing direction of the lens.

· The fastest pinhole video system is a mini-lens coupled to the camera. This is the best choice where the lowest cost and highest light efficiency are desired.

· A manual-iris lens is sufficient in applications where there are no large variations in light level, or where the light level can be controlled. Depending on the camera used, where there is more than a 50:1 change in light level, an automatic-iris pinhole lens or an electronically shuttered camera is needed.

· Most applications are solved using an “all-lens” system. In special cases where a thick barrier exists between a surface and the camera location, a rigid coherent fiber-optic bundle lens or bore-scope is used. If sufficient light

is available, an “all-lens” bore-scope type should be used to obtain the cleanest picture. Another alternative is a remote-head camera.

· AC power is preferred for permanent covert camera installations. Either 117 VAC to 12 VDC or 24 VAC wall-mounted converters are used. Using 12 VDC or 24 VAC is preferred over 117 VAC since it eliminates any fire or shock hazard and can be installed by security per-sonnel without outside help. Since most small cameras operate from 12 VDC, a 117 VAC to 12 VDC converter is most popular. For temporary installations, 12 VDC battery operation is used, with rechargeable or non-rechargeable batteries, depending on the application (Chapter 23).

18.10 SUMMARY

Pinhole lenses are used for surveillance problems that cannot be solved adequately using standard FFL or zoom lenses. The fast f-numbers of some of these pinhole lenses make it possible to provide covert surveillance under nor-mal or dimly lighted conditions. The small size of the front lens and barrel permit them to be covertly installed for surveillance applications.

A large variety of mini-lenses and pinhole lenses are available for use in covert security applications. These lenses have FL ranges from 3.8 to 22 mm covering FOVs from 12 to 95 . Variations, including manual- and automatic-iris, standard pinhole, mini- and off-axis-mini, provide the user with a large selection.
Equipment is available to provide covert surveillance under lighted or unlighted conditions. Through the use of IR illumination, scenes can be viewed in total darkness. Compact lenses, small and low-power cameras, wireless RF, microwave, and IR transmission systems make the covert system portable.

The availability of digital IP cameras has now made remote covert video surveillance a reality. The images from these cameras can be viewed using an Internet browser from any Internet access location by anyone having the camera IP address.