Attachment_2_-_SOW_Appendix_1_-_Chapter_4_UFC_4-021-02_CCTV_Systems.pdf

UFC 4-021-02

1 October 2013

CHAPTER 4 CLOSED CIRCUIT TELEVISION SYSTEMS

4-1 OVERVIEW.

The CCTV system is another core subsystem of an overall ESS. It is the collection of cameras, recorders, switches, keyboards, and monitors that allow viewing and recording of security events. The CCTV system can be integrated with ACS and IDS and may be centrally monitored at the Dispatch Center or locally monitored by security personnel at an individual facility. Uses of CCTV systems for security services include several different functions as described below.

4-1.1 Alarm Assessment.

When alerted by an alarm notification, CCTV cameras allow security personnel to visually assess the situation and make a determination as to what type of response may or may not be required. An example would be an intrusion alarm at a remote facility.

Visual assessment and other confirmation may indicate an unannounced maintenance crew at work. Symptoms of intrusion would lead to a response.

4-1.2 Access Control.

Cameras can be used by security personnel to visually identify persons and vehicles requesting entry prior to releasing a controlled portal (door, turnstile, gate, vehicle barrier, etc.).

4-1.3 Surveillance.

CCTV cameras can be used to give security personnel the capability to be made aware of or view visual events at multiple locations from a centralized remote viewing area.

CCTV camera technology makes visual information available that would normally only be available through multiple (possibly roving) human resources. Video analytics can significantly enhance surveillance effectiveness by cuing scenes of interest and highlighting areas within scenes for priority viewing by an operator.

4-1.4 Evidentiary Archives.

Retrieval of archived images may be helpful in identification or prosecution of trespassers, vandals, or other intruders.

As shown in Figure 4-1, a large networked CCTV system may encompass several buildings and can include multiple workstations where live and recorded video can be viewed. Both analog and Internet Protocol (IP) cameras can be used to gather video images which are then stored in a digital format. Building 1 represents a closed system with no connection to an outside network, in which case all video recording, viewing and management is done within the building.

Figure 4-1. Example Block Diagram for a Networked CCTV System.

4-2 CAMERAS.

Selecting the appropriate cameras is critical to a CCTV design. The following paragraphs provide guidance regarding basic camera types and features.

4-2.1 Color Versus Black and White.

Color cameras offer more information such as color of a vehicle or subject’s clothing.

Some ultra-low light color cameras are able to automatically sense the ambient light conditions and switch from color to black and white in low light conditions. Cameras must have auto-white balance to adjust for the changing color temperature of daylight and artificial lighting needed for night-time viewing. Black and white cameras are more sensitive than color cameras under low-light conditions and are best used when IR illuminators are required. These cameras are further described in the Viewing in Low- Light Conditions Section of this chapter.

Color cameras require a higher illumination level than black and white cameras to be effective. Typically, a high-quality color camera will work well down to 1 foot-candle (fc) (10 lux) of scene illumination, whereas a black and white camera might only require 0.1 fc (1 lux). These lighting level requirements vary with the camera model and manufacturer, so be sure to specify the appropriate illumination level for the scene of interest.

4-2.2 Indoor Cameras.

Indoor camera installations reduce the complexity of the system, but care must be taken to correctly specify the lens, field-of-view and camera hardware. Indoor cameras need:

• Sturdy, secure mounting.

• Auto-iris for lighting control.

• Auto-white balance to ensure proper color correction to accommodate changes in color temperature of lighting if it is dimmed or lighting is changed due to a light outage.

• To be mounted in a position to prevent glare from overhead lighting.

4-2.3 Outdoor Cameras.

Outdoor camera installations cost more than indoor cameras due to the need to environmentally house, heat, and ventilate the outside camera. When mounting a camera outdoors, the lighting requirement changes depending on the time of day and the weather. Because of this, consider the following for outdoor cameras:

a. Shrubs, trees, and other vegetation in a camera’s line of sight may cause obstructed views. Designers need to be aware of this when determining where to place cameras. Also, video motion detection systems can register a false positive when plants in the field-of-view move in windy conditions.

b. Provide heaters in cold weather applications.

c. Always use auto-iris lenses with outdoor cameras. The iris automatically adjusts the amount of light reaching the camera and thereby optimizes its performance. The iris also protects the image sensor from getting damaged by strong sunlight. Always set the focus in low light with an auto-iris lens. If the adjustment is made in sunlight, it is very easy to focus, but at night the iris diameter increases and the image is not in focus anymore. Special dark focus filters called “neutral density” filters or ND filters help reduce lighting by one or more stops of exposure. These filters do not affect the color of the image.

d. Use caution when mounting a camera behind glass. If you mount a camera behind glass, such as in housing, make sure that the lens is close to the glass. If the lens is too far away from the glass, reflections from the camera and the background will appear in the image.

e. Always try to avoid direct sunlight in an image. Direct sunlight blinds the camera and may permanently bleach the small color filters on the sensor chip, causing stripes in the image.

f. When using a camera outdoors, avoid viewing too much sky. Due to the large contrast, the camera will adjust to achieve a good light level for the sky, and the landscape and objects that must be assessed might appear too dark. One way to avoid these problems is to mount the camera high above ground. Use a pole if needed. Given mounting choices, mount cameras facing away from rising or setting sun, realizing that this varies by season.

g. Always use sturdy mounting equipment to avoid vibrations caused by strong wind. This is especially important with a long focal length lens.

These lenses amplify even the smallest movement of the mount. Building mounts are generally more stable than pole mounts. When in extremely windy conditions for a critical camera, consider using a gyro-stabilized mount lens to avoid vibration caused by wind. The gyro-stabilized lens has a cost premium and is not appropriate for general applications.

4-2.4 Fixed Position Cameras.

After being mounted, aimed, and focused by an installer, a fixed position camera provides a field of view that cannot be changed via remote control. When used for visually assessing intrusion or access control alarms, fixed cameras are good for review of pre-alarm conditions because there is a constant view of the scene in which the alarm was triggered. Pre-alarm allows the review of video information for the time period (typically ten to fifteen seconds) immediately before the alarm occurred. Pre-alarm video is often the most useful video content for determining the actual cause of the alarm.

Because of their static field of view, fixed cameras are well suited for video motion detection, but are not able to track a target of interest after it leaves the camera scene.

The installation and cost of fixed cameras is lower because there is no associated motor and control wiring.

4-2.5 Pan/Tilt/Zoom (PTZ) Cameras.

A PTZ camera contains a motorized mechanism for adjusting camera aim point and lens focal length, thus allowing an operator to dynamically change the field of view via remote control. This gives the operator a much better view of the overall area compared to a fixed camera. PTZ cameras are often used for both alarm assessment and video surveillance applications; however, they are not well-suited for pre-alarm assessment because they may not be focused on the alarm area at all times. Because of the drive motor, housing, and wiring for controls, PTZ cameras are typically three to four times more expensive than fixed cameras. Table 4-1 compares other salient parameters of fixed and PTZ cameras.

A PTZ camera can be controlled by an operator or it can be programmed to perform a guard tour during which it moves sequentially through a series of user-defined preset views. When not under operator or guard tour control, a PTZ can be set to return to a home position preset corresponding to the most important scene of interest. Preset views for alarm conditions can be programmed to override operator control, guard tour, and home position.

Table 4-1. Fixed versus PTZ Cameras.

Applications Cost Pre-alarm Review

Video Motion Detection

Intruder Tracking Capability

Fixed Alarm assessment for doors, gates and fence lines

Lower Recommend ed

Recommend ed

None

PTZ Surveillance for large open areas such as ports and airfields

Three times more expensive than a fixed camera

Poor application

Only for fixed, preset scenes

Good

4-2.6 Dome Cameras.

A dome camera is mounted in a hardened plastic lower-dome, which is commonly smoked-colored to conceal the camera. The use of smoke-colored domes provides covert lens positioning, while the use of clear domes provides for better low-light performance. Dome cameras are a good design solution for applications where the camera needs to be protected from the environment (such as dust) or it is desired to conceal the camera’s aim point. The variety of dome cameras is extensive, giving the designer a dome option for nearly any security application: fixed or PTZ, indoor or outdoor, full-size or mini-dome, analog or IP. A common application of dome cameras is in office buildings with suspended ceilings where aesthetics and ease of installation are important factors. PTZ dome cameras can move quickly from a home position to any preset, typically in less than two seconds.

4-2.7 IP and Analog Cameras.

A CCTV camera can be specified as either IP or analog, depending on the required video output format. As illustrated previously in Figure 4-1, an IP camera connects directly to an Ethernet switch from which the camera signal can be transmitted to any network node for viewing or recording. An analog camera typically connects to a digital video recorder from which live or recorded video can be transmitted to one or more CCTV workstations on the network. To ensure reliable video storage and viewing, IP cameras generally require a network with high bandwidth, high availability, and low latency.

4-3 ILLUMINATION.

4-3.1 Illuminance.

The CCTV designer must coordinate with the project’s lighting engineer, landscape architect and interior designer to ensure that illuminance within scenes of interest is sufficient for cameras to render full video. Meeting this objective involves analyzing two parameters - faceplate illuminance and scene illuminance - which are illustrated in Figure 4-2 and related by the following equation:

𝐂 = 𝐁𝐑�

𝐓 𝟒𝐍𝟐� where

C = faceplate illuminance (units are foot-candles or lux)

B = scene illuminance (units are foot-candles or lux)

R = scene reflectivity factor (dimensionless number between 0 and 1)

T = lens transmittance efficiency (dimensionless number, typical value is 0.8)

N = lens f-number (ratio of lens focal length to aperture diameter)

To illustrate the use of this equation, consider the following example:

A specific outdoor fixed camera has been proposed for use at an aircraft parking area.

Examining the manufacturer’s data reveals that the camera requires 0.0007 foot-candle of illuminance at the faceplate to generate useable video. To achieve the desired field of view, a 5-mm lens with an f-number of 1.6 and transmittance efficiency of 0.8 is proposed. During a nighttime lighting survey it is determined that the scene of interest includes dark-colored rotary wing aircraft parked on asphalt concrete. The reflectivity factor for this scene is estimated to be 0.07. Noting that existing light fixtures are in place, light meter readings are taken at several locations within the scene and the average illuminance value is calculated to be 1.2 foot-candle. Using this average scene illuminance, faceplate illuminance is calculated as follows:

𝐶 = (1.2)(0.07) � 0.8

(4)(1.6)2� = 0.007 𝑓𝑜𝑜𝑡𝑐𝑎𝑛𝑑𝑙𝑒𝑠

Comparing this calculated value with the manufacturer’s specification (0.0007 foot-candle) indicates that existing nighttime illumination at the aircraft parking area is more than adequate to support full video capture by the proposed camera.

4-3.2 Uniformity.

Uniform illuminance within a camera scene yields the highest video quality. While not always achievable, the designer should strive for an average-to-minimum uniformity ratio of 4:1 within a scene of interest. Video quality degrades noticeably when the uniformity ratio exceeds 8:1. For additional guidance on uniformity and other CCTV illumination parameters, refer to UFC 3-530-01.

Figure 4-2. Scene Illuminance and Faceplate Illuminance.

Reflectivity factors for a range of surface conditions are presented in Table 4-2, and these values can be used to estimate reflectivity for actual scenes of interest.

Table 4-2. Reflectivity Factors for Various Surface Conditions.

Scene Description Reflectivity Factor

Asphalt concrete 0.07

Grass and trees 0.2

Red brick 0.35

Portland cement concrete 0.4

White matte painted surface 0.6

Glass window or wall 0.7

Snow-covered surface 0.85

4-3.3 Glare Reduction.

Glare is very detrimental to CCTV camera performance as illustrated in Figure 4-3.

Glare reduction can be achieved by specifying full cut-off luminaires and insuring that luminaires are not in a camera’s field of view. In general, the source of illumination is best located above the level of the camera. The CCTV designer must coordinate with the lighting designer to ensure that these glare-reduction objectives are met.

Figure 4-3. Effect of Glare on CCTV Camera Image Quality.

4-3.4 Interior Lighting.

Interior lighting for CCTV presents special issues that need to be considered by the designer. For example, after-hours lighting may be significantly lower than normal operation lighting. Two solutions help minimize this impact.

a. The first technique is the use of cameras with automatic backlight compensation. Backlight compensation is a camera feature that enables the camera to automatically adjust picture brightness depending on lighting conditions, which compensates for bright backgrounds so foreground objects are not silhouetted. Frequently, CCTV cameras near windows are affected by backlighting, causing shadows and silhouettes, so the use of appropriate cameras with backlight compensation is effective.

b. The second technique is the use of cameras with automatic gain control, a feature that amplifies existing video to help a camera create an enhanced video signal at low light levels.

Both of these techniques enable cameras to function more effectively in interior low-light conditions and are useful for outdoor cameras as well. In some cases, the integration of CCTV cameras with night-lights and intrusion sensors can be very effective. The sequence of events might be as follows: an intruder activates an interior presence sensor which, in turn, activates instant-on lighting for CCTV camera assessment.

4-4 VIEWING IN LOW-LIGHT CONDITIONS.

In addition to increasing the illumination level of the surrounding area, several technology solutions are available to permit viewing under low-light conditions. These include black/white switching cameras, infrared illuminators, or thermal imagers. These technologies are often used where visible light either brings undesired attention to a critical facility, or surrounding property owners object to visible light adequate for good visual camera operation.

4-4.1 Black/White Switching.

Many cameras will automatically switch from color during daytime to black/white at night, which permits viewing under low light conditions. This can be an effective solution in situations where the existing illumination levels are too low during night conditions to permit color camera use, but color camera use is desired during daytime conditions.

Numerous CCTV camera manufacturers offer auto-switching black/white cameras.

4-4.2 Infrared Illuminators.

The human eye cannot see infrared light. Most monochrome CCTV (black/white) cameras, however, can. Thus, invisible infrared light from either an LED or laser source can be used to illuminate a scene, which allows night surveillance without the need for additional artificial lighting. IR illumination patterns can be matched to camera field of view. A variety of patterns are available ranging from narrow- to wide-angle and short-to long-range coverage. LED IR illuminators are a good choice for short-range flood coverage and medium-range spot coverage. Approximate coverage ranges for a 26-watt LED illuminator are 65 feet (20 m) for a 120-degree flood pattern and 310 feet (95

m) for a 10-degree spot pattern. Laser IR illuminators should be considered when the desired coverage range exceeds the capabilities of LED illuminators. For example, a 60-watt laser IR illuminator can project a 10-degree pattern to a maximum effective distance of approximately 2300 feet (700 m). Infrared provides the following benefits over conventional lighting:

• Extended service life - up to 10 years.

• Lower running costs (but higher installation costs).

• Covert surveillance - no visible lighting to alert or annoy neighbors.

It is important to design illumination specifically for the CCTV camera being used. For example, infrared illuminators require black/white cameras and do not work on color cameras, unless the color camera has an automatic black-and-white switching feature.

Cameras will not render color images when used under infrared illumination. The range that the camera will see in the dark depends on sensitivity and spectral response of the camera and lens combination. Many black and white cameras use infrared filters to intentionally filter out non-visible light. Therefore, black and white cameras which are designed to be used in conjunction with infrared lighting must not have an infrared filter.

Dual mode cameras that can switch from color to monochrome operation in low light conditions must not have an infrared filter for the reason cited above.

4-4.3 Thermal Imagers.

Thermal imagers use a special technology that senses heat signatures rather than visual information. These cameras operate under complete darkness. Thermal imagers are best used in long-range detection and surveillance applications. Thermal imagers detect and display images based on infrared energy emitted from objects rather than visible light reflected off objects. The most common technology is Forward Looking Infrared (FLIR). Thermal cameras work on a temperature differential between the object and the background. In desert environments, the background is white and people are black. In cooler environments, the background is black and people are shown as white images. A key advantage of long-range thermal imagers is that they are less susceptible to environmental influences from rain and fog in comparison to visible light cameras. The disadvantage of thermal-imagers is the high cost and the inability to discern facial features and other fine details in the scene.

Typically thermal imagers are classified as medium or long wavelength as illustrated in Table 4-3. For security applications in which the object of interest is a man-size target within a few hundred meters of the camera, uncooled long-wavelength imagers are preferred because of their lower cost, both in terms of initial purchase and lifecycle maintenance. Cooled medium-wavelength imagers, though more costly, can resolve very small thermal gradients and, equipped with the appropriate lens, can capture images of man-size targets at ranges out to a few thousand meters.

Table 4-3. Characteristics of Thermal Imagers.

Classification Wavelength

Cooling Cost Recommended Service Period

Medium Wavelength

3- 5 microns Cryogenically cooled

$50K - $150K 7,500 hours

Long Wavelength

7-14 microns Uncooled $5K - $50K 30,000 hours

4-5 ANGLE OF VIEW AND FIELD OF VIEW.

An important consideration when designing a CCTV system is determining the desired field of view for each camera. Field of view and angle of view are illustrated in Figure 4-

4. Based on the desired field of view, the designer must specify the appropriate camera mounting location, aim point, format, and lens focal length to capture the required image.

Figure 4-4. Angle of View and Field of View.

Camera format refers to the nominal diagonal measure of the image sensor (also called the faceplate), and typical sizes are shown in Table 4-4. Given the faceplate dimensions and the lens focal length, the angle of view for any camera-lens combination can be calculated as follows:

𝜶 = 2 𝒕𝒂𝒏−𝟏 𝒍 2f where α = angle of view (horizontal or vertical) 𝑙 = length of image sensor (width or height, refer to Table 4-4) f = focal length of lens

A relatively narrow angle of view would be provided by a 2/3-inch format camera equipped with 50-mm lens. In this example, the angles of view would be as follows:

𝜶𝒉𝒐𝒓𝒊𝒛𝒐𝒏𝒕𝒂𝒍 = 2 𝒕𝒂𝒏−𝟏 � 𝟖.𝟖

(𝟐)(𝟓𝟎)� = 𝟏𝟎.𝟏° 𝜶𝒗𝒆𝒓𝒕𝒊𝒄𝒂𝒍 = 2 𝒕𝒂𝒏−𝟏 � 𝟔.𝟔

(𝟐)(𝟓𝟎)� = 𝟕.𝟔°

A relatively wide angle of view would be provided by a 1/3-inch format camera equipped with 5-mm lens. In this example, the angles of view would be as follows:

𝜶𝒉𝒐𝒓𝒊𝒛𝒐𝒏𝒕𝒂𝒍 = 2 𝒕𝒂𝒏−𝟏 � 𝟒.𝟖

(𝟐)(𝟓)� = 𝟓𝟏.𝟑° 𝜶𝒗𝒆𝒓𝒕𝒊𝒄𝒂𝒍 = 2 𝒕𝒂𝒏−𝟏 � 𝟑.𝟔

(𝟐)(𝟓)� = 𝟑𝟗.𝟔°

The following simple ratios can be used to perform several types of field of view calculations:

𝒍𝒘𝒊𝒅𝒕𝒉 𝑾 𝒍𝒉𝒆𝒊𝒈𝒉𝒕 𝑯 𝒇 𝑫 lwidth = width of image sensor W = width of field lheight = height of image sensor H = height of field 𝒇 = focal length of lens D = distance from lens To show the application of these field-of-view ratios, consider this example. A ½-inch format camera will be mounted on a pole located 30 feet (10 m) from a pedestrian turnstile on the perimeter of an outdoor restricted area. If the desired vertical field of view is a full-height image of each person entering the restricted area, what lens focal length is needed? Assuming a 7-foot (2,134 mm) height of field at a distance of 30 feet (9,144 mm), lens focal length can be calculated as follows:

𝒇 = 𝑫𝒍𝒉𝒆𝒊𝒈𝒉𝒕 𝑯

= (𝟗𝟏𝟒𝟒)(𝟒.𝟖) 𝟐𝟏𝟑𝟒

= 20.6 mm

In this example, the designer could specify a varifocal lens to allow some fine tuning of the field of view during the installation process or later adjustment to meet a new field-of-view requirement. A 10-40 mm varifocal lens set at 20.6 mm would meet the stated objective of providing a full-height image of a person at the turnstile, but the same varifocal lens could be set to 40 mm to provide better facial detail or to 10 mm to cover an adjacent vehicle gate. Supporting calculations for these focal lengths are as follows:

𝑯 = 𝑫𝒍𝒉𝒆𝒊𝒈𝒉𝒕 𝒇

= (𝟗𝟏𝟒𝟒)(𝟒.𝟖) 𝟒𝟎

= 𝟏,𝟎𝟗𝟕 𝒎𝒎 = 3.6 feet (Better Facial Detail)

𝑾 = 𝑫𝒍𝒘𝒊𝒅𝒕𝒉 𝒇

= (𝟗𝟏𝟒𝟒)(𝟔.𝟒) 𝟏𝟎

= 𝟓,𝟖𝟓𝟐 𝒎𝒎 = 19 feet (Cover Vehicle Gate)

Table 4-4. Typical Faceplate Sizes.

Nominal Diagonal Measure

Actual Width Actual Height

1/4 inch 3.2 mm 2.4 mm 1/3 inch 4.8 mm 3.6 mm 1/2 inch 6.4 mm 4.8 mm 2/3 inch 8.8 mm 6.6 mm 1 inch 12.8 mm 9.6 mm

4-6 CAMERA RESOLUTION.

Camera resolution refers to the “graininess” of images captured and transmitted by a camera and is expressed in terms of televisions lines (TVL) for analog cameras and picture elements (pixels) for IP cameras. Table 4-5 correlates qualitative resolution descriptions to equivalent camera specifications for both analog and IP cameras. For each scene of interest, the CCTV designer must determine the camera resolution required to achieve the desired discrimination level for objects in the scene. Visual target discrimination criteria developed by John Johnson in the 1950’s, commonly referred to as the Johnson criteria and summarized in Table 4-6, can be used to analyze the impact of camera resolution on object discrimination for any given angle of view. These criteria are based on a 50% probability of accurate discrimination by a person viewing the camera image. The following example applies the Johnson criteria to illustrate the difference in object discrimination performance between a high-resolution camera and a megapixel camera.

Table 4-5. Typical Camera Resolution Specifications.

Qualitative Description

Equivalent Camera Resolution Specification Analog Camera IP Camera

Very Low Resolution - Quarter VGA N/A 320 X 240 pixels

Low Resolution - VHS 330 TVL N/A

High Resolution - DVD 540 TVL 720 X 480 pixels

Megapixel Resolution N/A 1280 X 1024 pixels (and higher)

HD 1080p Resolution N/A 1920 x 1080 pixels

Table 4-6. Object Discrimination Levels Based on Johnson Criteria.

Discrimination Level Meaning

Pixels required across minimum dimension Example

Detection An object of a specified size is present.

2 An object with minimum dimension of 60 inches in the vertical orientation is present in the scene.

Recognition The class to which the object belongs can be determined.

8 The object is a vehicle, not people or animals.

Identification The type of object within the class can be determined.

16 The vehicle is a sedan, not a truck, SUV or van.

A 2/3-inch format camera with a 25-mm lens will be used to visually assess outdoor perimeter intrusion alarms along a restricted area. The camera will be aimed parallel to the fence line to view objects in the clear zone. The objective is for the camera to provide recognition level discrimination for a person crawling through the clear zone.

This level of discrimination will allow a human crawler to be distinguished from objects of similar size such as animals and wind-blown debris. The minimum dimension for a human crawler is 12 vertical inches (0.3 m), and applying the Johnson criteria suggests that this 12-inch (0.3 m) vertical profile must be “painted” by 8 vertical pixels on the image collected by the camera. Given the properties of the camera, lens, and object to be viewed, the following equation can be used to calculate the maximum range for “recognizing” the human crawler:

𝑫 = 𝒉𝒇𝑹𝒗𝒆𝒓𝒕𝒊𝒄𝒂𝒍 𝒑𝒗𝒆𝒓𝒕𝒊𝒄𝒂𝒍𝒍𝒉𝒆𝒊𝒈𝒉𝒕

D = distance from lens to object h = height of object f = focal length of lens

Rvertical = vertical resolution of the camera in pixels pvertical = vertical pixels required on object lheight = height of image sensor

Converting the height of the human crawler from 12 inches to 305 mm and solving this equation yields the following result for a high-resolution (720 X 480 pixels) camera:

𝑫 = (𝟑𝟎𝟓)(𝟐𝟓)(𝟒𝟖𝟎)

(𝟖)(𝟔.𝟔) = 𝟔𝟗,𝟑𝟏𝟖 𝒎𝒎 = 𝟐𝟐𝟕 𝒇𝒆𝒆𝒕

A megapixel camera (1,280 X 1,024 pixels), by comparison, provides a maximum “recognition” range of:

𝑫 = (𝟑𝟎𝟓)(𝟐𝟓)(𝟏,𝟎𝟐𝟒)

(𝟖)(𝟔.𝟔) = 𝟏𝟒𝟕,𝟖𝟕𝟗 𝒎𝒎 = 𝟒𝟖𝟓 𝒇𝒆𝒆𝒕

In this illustration, even though the angles of view are the same for both cameras (20° horizontal & 15° vertical), the megapixel camera has two times the effective range of the high-resolution camera.

For objects having a minimum dimension equal to their width (such as a person walking), the following equation can be used in conjunction with the Johnson criteria to calculate maximum effective range:

𝑫 = 𝒘𝒇𝑹𝒉𝒐𝒓𝒊𝒛𝒐𝒏𝒕𝒂𝒍 𝒑𝒉𝒐𝒓𝒊𝒛𝒐𝒏𝒕𝒂𝒍𝒍𝒘𝒊𝒅𝒕𝒉

D = distance from lens to object w = width of object f = focal length of lens Rhorizontal = horizontal resolution of the camera in pixels phorizontal = horizontal pixels required on object lwidth = width of image sensor

4-7 VIDEO FRAME RATE.

Video frame rate is an important CCTV design parameter that affects video transmission, storage, and display. A frame rate of 30 frames per second (fps) is generally considered to be “full motion video” based on the National Television Standards Committee (NTSC) analog video standard. However, the transmission and recording frame rates for digital video can be set for a range of values between 1 and 30 fps, and some IP cameras and network video recorders support frame rates up to 60 fps. For most security applications, including alarm assessment and evidentiary archives, frame rates between 4 and 10 fps are fully adequate. Higher frame rates are appropriate for surveillance applications in which a smooth video stream is beneficial to the operator, especially when using a PTZ camera. Frame rates of 24 fps and higher will be perceived as “smooth” when viewed by an operator for extended periods of time.

In most digital video systems, transmission and recording frame rates can be programmed to automatically change in response to an external event such as an intrusion alarm or an internal video motion detection trigger.

4-8 DIGITAL VIDEO BANDWIDTH.

The CCTV designer must coordinate closely with the appropriate network designer or administrator to ensure that network bandwidth will support digital video transmission.

Estimating bandwidth requirements using the equation below will aid in this coordination.

𝒃 = 𝒌𝒓𝒛 where b = bandwidth required for a single video stream k = network overhead factor, typical value is 1.4 r = video frame rate z = average compressed file size of a single video frame

The average file size for a single frame is dictated by the video resolution and compression, and typical values are given in Table 4-7. The following example illustrates the use of this table and the equation above.

Four IP cameras will be installed in an administrative building as part of an access control upgrade. One fixed camera will be installed at the main entry door and the other three fixed cameras will be installed at doors designated for emergency exit only. Each camera has megapixel (1280 X1024) resolution and H.264 video compression. Video from the administrative building will be transmitted via network at 5 fps to a headquarters building for recording and viewing. How much network bandwidth is required to support these four cameras?

Assuming that the main entry camera will have high scene activity, a single video frame will have an average compressed file size of 72 kB. The other three cameras will each have low scene activity and a corresponding file size of 36 kB. Using these file size values from Table 4-7, along with the specified frame rate of 5 fps, the following bandwidth estimates can be made, with the results expressed in units of Megabits per second (Mbps):

ENTRY DOOR CAMERA: 𝐛𝐞𝐧𝐭𝐫𝐲 = (𝟏.𝟒)(𝟓)(𝟕𝟐) = 𝟓𝟎𝟒 𝐤𝐁𝐩𝐬 = 𝟒.𝟎𝟑 𝐌𝐛𝐩𝐬

EXIT DOOR CAMERA: 𝐛𝐞𝐱𝐢𝐭 = (𝟏.𝟒)(𝟓)(𝟑𝟔) = 𝟐𝟓𝟐 𝐤𝐁𝐩𝐬 = 𝟐.𝟎𝟐 𝐌𝐛𝐩𝐬

TOTAL FOR ALL CAMERAS: 𝐛𝐭𝐨𝐭𝐚𝐥 = 𝐛𝐞𝐧𝐭𝐫𝐲 + 𝟑𝐛𝐞𝐱𝐢𝐭 = 𝟏𝟎.𝟎𝟗 𝐌𝐛𝐩𝐬

(Note: 1 Mbps = 125 kBps) Based on this calculation, the four IP cameras will require approximately 10 Mbps of network bandwidth to transmit video from the administrative building to the headquarters building.

Table 4-7. Single-frame File Size for Various Resolution Values and Compression Schemes

Resolution (H x V pixels)

Average Compressed File Size (kB) for a Single Video Frame H.264 - Low Scene

Activity H.264 - High Scene

Activity MJPEG - High

Quality 320 X 240 4 8 20

720 X 480 12 24 60

1280 X 1024 36 72 180

1920 X 1080 50 100 250

4-9 DIGITAL VIDEO RECORDING.

The CCTV designer must plan for digitally recording video from all cameras, and several technology options are available. The following paragraphs provide a brief overview of four of the most common video recording methods followed by an explanation of how to estimate video storage requirements.

4-9.1 Memory Card.

Several camera models have a built-in memory card which allows video to be recorded at the edge of the CCTV system. This ensures uninterrupted recording even when the connection between the camera and the central system is lost, but storage capacity is very limited. An SD card, for example, has a capacity of 2 GB.

4-9.2 Digital Video Recorder (DVR).

A DVR digitizes analog camera inputs and stores the video on one or more internal hard drives. The number of camera inputs ranges from 4 to 32, depending on the model selected. Recording resolution up to 720 X 480 at 30 fps is available on many models.

Access to live and recorded video from remote workstations is enabled by a network interface card in the DVR, but video will continue to be recorded even if network connectivity is lost. A DVR can provide several terabytes of video storage capacity.

4-9.3 Network Video Recorder (NVR).

An NVR records digital video from multiple IP cameras and video encoders to one or more internal hard drives. The capacity and sophistication varies greatly within this category of video recording technology. A low-end NVR can typically record up to 16 cameras at 30 fps, with recording resolution no greater than 720 X 480. A high-end NVR can accommodate 50 or more cameras, recording each camera at 24 fps with HD 1080p resolution. All NVRs have an internal network interface card to receive and distribute IP video streams, and high-end units have two or more Gigabit Ethernet ports.

A CCTV designer must ensure that each network path connecting an IP camera with its associated recording node is adequate in terms of both availability and bandwidth. NVR storage capacity ranges from 1 TB at the low end of the category to greater than 10 TB for high-end units.

4-9.4 Hybrid Video Recorder (HVR).

Combining the functionality of a DVR and an NVR into a single unit, an HVR can digitize and record multiple analog camera inputs while simultaneously recording multiple IP video streams. Many HVRs provide up to HD 1080p recording resolution for IP cameras, and HVR video storage capacities generally range from 2 to 10 TB.

4-9.5 Required Storage Capacity.

Once all cameras have been specified for a project, the CCTV designer should use the following equation to estimate the required storage capacity for each video storage device:

𝐬 = 𝐭𝐫𝐳 where s = required video storage capacity for a single camera t = required video storage duration r = video frame rate z = average compressed file size of a single video frame

The following example illustrates the methodology and calculations needed to estimate required video storage capacity for a single storage device.

The CCTV system for an access control point will be upgraded. All existing components, with the exception of two analog PTZ cameras, will be removed and replaced with 5 new megapixel IP cameras, all fixed, and an NVR. Video encoders will be used to convert the analog camera signals to IP video streams which will be recorded to the NVR along with the 5 IP camera feeds. The security manager has stated that the fixed cameras will be recorded at 4 fps with a resolution of 1280 X 1024, and the PTZ cameras will be recorded at 10 fps with a resolution of 720 X 480. The security manager also stated that there is a 7-day video storage requirement for all access control point cameras. The IP cameras and the encoders for the PTZ cameras will all use MJPEG compression. How much video storage is required for the NVR?

Converting 7 days to 604,800 seconds, the storage required for a single fixed IP camera can be calculated as follows:

𝐬𝐟𝐢𝐱𝐞𝐝 = (𝟔𝟎𝟒,𝟖𝟎𝟎 𝐬)(𝟒 𝐟𝐩𝐬)(𝟏𝟖𝟎 𝐤𝐁) = 𝟒𝟑𝟓,𝟒𝟓𝟔,𝟎𝟎𝟎 𝐤𝐁 ≅ 𝟒𝟏𝟓 𝐆𝐁

The storage required for a single PTZ camera can be calculated in a similar manner:

𝐬𝐩𝐭𝐳 = (𝟔𝟎𝟒,𝟖𝟎𝟎 𝐬)(𝟏𝟎 𝐟𝐩𝐬)(𝟔𝟎 𝐤𝐁) = 𝟑𝟔𝟐,𝟖𝟖𝟎,𝟎𝟎𝟎 𝐤𝐁 ≅ 𝟑𝟒𝟔 𝐆𝐁

Taking into account the quantity of fixed and PTZ cameras, the total storage required for all cameras can be calculated as follows:

𝐬𝐭𝐨𝐭𝐚𝐥 = (𝟓)(𝟒𝟏𝟓 𝐆𝐁) + (𝟐)(𝟑𝟒𝟔 𝐆𝐁) = 𝟐,𝟕𝟔𝟕 𝐆𝐁 ≅ 𝟐.𝟕 𝐓𝐁

(Note: 1 TB = 1024 GB = 1,073,740,000 kB)

4-10 CCTV WORKSTATION.

To allow viewing of live and recorded video, the designer must specify at least one workstation for each CCTV system, and multiple workstations may be required for a large distributed system. Because of the computational demands associated with processing and displaying digital video streams, a “gaming” computer is a good choice for a CCTV workstation. These computers generally have high-speed processors, large amounts of RAM, fast graphics cards with high-resolution output, and network interface cards supporting Gigabit Ethernet speed. For a single-operator workstation, a graphics card feeding one or two monitors will usually be sufficient for CCTV viewing and management. If three or four monitors are needed for a workstation, two dual output graphics cards or a single quad output graphics card must be specified for the workstation. The graphics card and monitor must provide display resolution equal to or greater than the highest resolution camera in the system. Any workstation that will be used to control PTZ cameras must be equipped with a joystick. Video management software that is compatible with all cameras, encoders, and recording devices must be installed on each workstation.

4-11 VIDEO ANALYTICS.

If surveillance is an important security objective for a CCTV system, the designer must consider including video analytics as part of the system specification. Video analytics software allows the user to input a specific set of rules for each scene of interest, which, if violated, generate visual cues on the monitor, thus drawing the operator’s attention to suspicious objects or behaviors. This capability to automatically prioritize scenes and highlight suspicious areas for the operator maximizes the effectiveness of surveillance activities. Video analytics can be especially beneficial when a single operator is required to perform surveillance with a large number of cameras. Video analytics algorithms can be embedded in IP cameras, encoders, and recording devices or they can run on dedicated file servers. Common rule violations programmed to alert the operator include crossing a virtual tripwire, loitering in a prohibited area, moving in the wrong direction, leaving an unattended object, and removing an object.

4-12 CCTV DESIGN PROCESS SUMMARY.

4-12.1 Define Security Objectives for the CCTV System.

Begin by evaluating specific project requirements in light of the four most common CCTV functions: 1) alarm assessment, 2) access control, 3) surveillance, and 4) evidentiary archives. Concisely state objectives with enough detail to facilitate camera selection and layout. Example objectives are as follows:

• Visually assess perimeter intrusion alarms for seven bistatic microwave sensor zones around the satellite communications facility.

• Visually identify the driver and vehicle prior to opening the gate at the test area.

• Perform surveillance of four exhibit areas in the museum and maintain a 30-day video archive for evidentiary purposes.

4-12.2 Develop a Camera Layout to Meet the Security Objectives.

Indicate camera locations on site plans and building floor plans, identifying each camera as fixed or PTZ. Specify the mounting configuration (wall, ceiling, pole, roof, etc.) for each camera, and select the appropriate camera and lens for the intended field of view.

4-12.3 Verify That Illumination Is Sufficient For Each Scene Of Interest.

Ensure that camera specifications for faceplate illumination are met and that uniformity ratios are within acceptable limits. Specify lighting upgrades as needed.

4-12.4 Specify Workstation Locations.

Indicate workstation locations on building floor plans, and describe the basic configuration of each workstation including quantity and size of monitors. Identify any special furniture or console requirements.

4-12.5 Specify Recording Locations and Capacity.

Indicate recording locations on building floor plans, and describe the type and quantity of recording devices required at each location. Calculate the required video storage capacity for each recording device.

4-12.6 Define Network Architecture.

Develop a block diagram to illustrate connectivity for all cameras, recorders, workstations, and networking devices. Specify cables required for equipment interconnection, and calculate bandwidth requirements for all network connections.

4-12.7 Define Power Requirements.

Determine the power requirements for each component. Specify all power circuits and the location of all power supplies.

4-12.8 Describe Software and Integration Requirements.

Specify features and functions required for camera control, video management, and analytics. State alarm assessment requirements for integration with intrusion detection and access control software.