1. Introduction

In this era of rapid technical advancement, the marketplace is flooded with
a tremendous variety of video equipment. This can be both good and bad.
It is good from the standpoint that a potential purchaser of video
equipment can almost certainly find a commercially available video unit
or package that will meet perceived needs and desires. However, when
those needs are not clear, or are changing, and the characteristics of the
equipment are not understood, the number of equipment choices can lead
the purchaser to a feeling of confusion or helplessness. Video equipment
salespeople can confound the issue further by advocating the purchase of
products which have attractive (but unnecessary) features and capabilities,
or by simply recommending the more well-known manufacturers'
products, which tend to be more expensive.

Personnel in the law enforcement and corrections agencies wishing to
utilize video surveillance systems (for collecting evidence and promoting
officer safety, for example) are challenged when required to confront these
equipment choices and sales pressures while staying within an established
budget. The purpose of this guide is to assist those law enforcement and
procurement officials who are not technically trained in video equipment
in the selection and application of video surveillance equipment that will
satisfy their needs. This guide primarily addresses general-use video
equipment, including separate video cameras, self-contained camcorders,
video recorders/players, and video display systems (monitors). However,
special-purpose video equipment is also described, such as the Patrol Car
Surveillance System.[1]

The guide begins with a discussion of typical video surveillance
assignments, that is, a definition of user requirements for the law
enforcement and corrections communities. This requirements definition
serves as a jumping-off point and reference base for all subsequent
deliberations in later sections of the guide. An overview of the available
video technology is presented next, along with a summary of tape formats.
A delineation of the technical parameters that most influence operational
performance for the various types of gear follows. Guidance is provided
regarding the application of specific types of video equipment to meet
functional requirements. Another important element of the guide is
information on the latest advancements in video technology and the effects
those advancements will have on surveillance work. With cost
information, the functional requirements data will help sort out the lowest
cost equipment that can effectively satisfy at least the minimally
acceptable surveillance requirements established by the law enforcement
and corrections community.

The appendices offer detailed experimental methods and results that are
summarized in the main text. This will assist advanced users in
determining what types of tests might be appropriate when evaluating new
equipment for use in their application environment.

--------------------------------

2. Video Surveillance Requirements

2.1 Typical Video Surveillance Assignments

In order to develop a guide that would be useful to the law enforcement
and corrections community, it was first necessary to determine the
community's video surveillance needs. This was accomplished through the
development of a survey to which state and local law enforcement
agencies responded. The survey focused on uncovering the kinds of video
surveillance assignments required of a typical police department. Those
assignments (and the particular users' needs to accomplish the
assignments) would lead naturally to the specifications for equipment.
Table 1 presents a set of representative surveillance assignments based on
results from the survey.

Besides the basic functional assignments, a complete description of video
surveillance applications must address performance under certain
operating conditions. One of the most obvious of these conditions is the
environment to which the equipment will be subjected. However, there are
other operating conditions that can affect the usefulness of the equipment
and the success of the surveillance assignment. They can be placed into
two broad categories - usage and power requirements. Table 2 contains a
list of environmental, usage and power conditions that can potentially
affect the performance of a surveillance system.

It is necessary to define in finer detail the scope of three of the most
important usage requirements: light, clarity, and distance. The range of
values for each usage condition will help determine the feasibility of using
certain types of video equipment and the minimum equipment
specifications required of them. For example, lighting levels must be
classified to reflect various indoor and outdoor settings.

Clarity, as a usage condition, is meant to relate how explicit the surveyed
image has to be to satisfy the intended use of the user. From a law
enforcement and corrections perspective, clarity can be subdivided into
two quality levels - being able to identify faces, and being able to identify
figures and activities. Distance classifications are related to the clarity of
images and to typical scenarios of police operations (i.e., where the police
are and where the subject is). Table 3 summarizes the subclassifications of
light, clarity, and distance conditions.

2.2 Other Survey Results

The survey revealed that there are several areas of interest to the law
enforcement and corrections agencies. The area that elicited the most
interest was still video, a newer technology that shows much promise for
the surveillance community. Agencies believe this will be particularly
useful in recording information at the scene of a crime, but it will also be
useful for video mug shots and forensic data collection. Also of great
interest was low-light, amplified-light and infrared video equipment for
use in night surveillance. This would be used primarily to provide a means
to track the movements of suspects, and to better perform building and
area surveillance. The third significant area of interest was identification
of suspects at distances greater than 200 yards. Where equipment is
concerned, slightly more agencies currently procure more general-purpose
equipment than specialized equipment. The most common piece of
equipment in use is a consumer quality (VHS, S-VHS, 8 mm, or Beta)
videocassette recorder. The other two most common pieces of equipment
in use are low to medium resolution color cameras and low to medium
resolution camcorders. The frequency of use of these types of equipment is
logical in that they are the easiest to obtain and use. However, they are not
the only type of equipment in use. The only piece of equipment that the
surveys did not indicate was in use was a camera concealed on a person. In
light of this survey result, body cameras are not included in this guide.

The physical treatment of video equipment can vary greatly. Many pieces
of equipment are subjected to vibration, moisture, and both heat and cold,
while others are only used in environmentally controlled areas. The field
storage conditions of the equipment also vary greatly. They range from
custom mounts in vehicles to car trunks and seats. The temperatures in
these surroundings can vary more than the usage (outside) conditions,
especially if the car is unoccupied for more than 30 min at a time. The
permanent storage space for the equipment, however, is fairly consistent:
usually an office environment or air-conditioned room.

Also of interest is the agencies' video equipment operators. Survey
responses indicate that most of the agencies have video specialists who are
the only ones to operate video equipment. There is also, however, a
significant number of responses from agencies where everyone is required
to operate at least some video equipment.

One area of possible concern is the amount of training available for the
video equipment operators. Only one agency reported that it had more than
2 percent of its training budget available for training on the mechanics and
techniques of video equipment usage. Several reported that no budget was
available for this type of training.

2.3 Summary

Video surveillance requirements for law enforcement and corrections span
a number of different applications but may be cataloged into a
fundamental set of six areas. These requirements areas are:

1. Identifying subjects (including persons) at varied distances and at varied
light levels. Different quality levels of identification are required for
different applications (e.g., positive facial identification of persons under
surveillance versus identification of persons breaking into a building to
prompt security forces to respond).

2. Recording/documenting data and/or evidence during or after a crime.
Exact color may be very important to immediately apprehend a suspect
based on what he/she was wearing, or it may be necessary to record
precisely the hue of mud at a crime scene.

3. Handling scenes/locations with multiple activities and/or multiple
subjects. The responsiveness of cameras or camcorders may make the
difference between capturing on film only one illegal act or several.

4. Covering indoor and outdoor activities in different geographic areas. No
two police departments deal with exactly the same environment.

5. Establishing multi-purpose flexibility so as to allow selected equipment
to operate with other existing or new equipment. Equipment suites require
physical and functional compatibility.

6. Promoting operational effectiveness. The best equipment in the world
will not produce results if it cannot be used by law enforcement and
corrections personnel effectively when it is required. Limitations on
training budgets make it mandatory that equipment operation also be
straightforward.

--------------------------------Order your surveillance system

3. Description of Video Surveillance Components and Systems

3.1 Overview

When selecting equipment for video surveillance applications, there are a
number of choices to make. This section reviews the four basic types of
video surveillance equipment - cameras, camcorders (camera-recorders),
recorders/players, and video displays (monitors/televisions) - that can be
used (in some combination) to form a complete video system. Cameras are
presented first. In the technology description, the difference between tube
and solid-state (i.e., charge coupled device [CCD]) camera units is
explained, as well as differences between analog and digital cameras and
camcorders. After a summary of video camera features and an outline of
lenses, two special camera types are described. Still-video cameras and
low-light cameras are addressed in the context of continuing and emerging
law enforcement applications.

Since magnetic tape is the primary storage medium used in video
equipment, a description of videotape technology and quality has been
included at the beginning of the discussion related to tape machines (i.e.,
camcorders and recorders/players). Where taping is concerned, there are
six video tape formats (and their derivatives) that are applicable to video
surveillance: D1, D5, DV, Betacam [trademark] (analog and digital), VHS,
and 8 mm. In many cases, the tape format and its inherent capabilities will
strongly influence equipment selection.

Any discussion of video surveillance equipment becomes confusing
immediately unless the equipment types (camcorders, etc.) are categorized
into smaller, similar groups based on quality levels. There are two primary
performance parameters that can be used to differentiate quality levels for
each video equipment type - resolution and color. Resolution, expressed
simply, is how clearly one can distinguish the detailed parts of an image.
Color relates to the ability of equipment to record, display, or playback
color or black and white images. These performance parameters, and other
equipment characteristics that directly influence the quality levels of video
equipment, will be explained fully in later sections of this guide.

Table 4 presents the breakdown of equipment types based on their quality
levels. Equipment types with low- and medium-resolution capability have
been lumped together (within a color or black and white category) because
they employ essentially the same technology. The equipment with the
greater (medium) resolution has taken advantage of later technological
refinements to achieve a higher quality level. High-resolution video gear
(also broken down by a color or black and white capability) was especially
planned and designed to accommodate high-quality applications, such as
commercial television production and broadcast. It is unlikely that low-
and medium-resolution equipment will ever migrate to the levels of the
high level gear through subsequent design refinements. As with most
electronic equipment, it will be shown later that improvements in video
equipment capability and quality result in differences in price; most
significantly as the equipment quality jumps from a medium-resolution
level to one of high-resolution.

3.2 Video Cameras

3.2.1 Technology Summary

Tube cameras have been around since the beginning of television and the
electronic video industry. There have been several names associated with
tube cameras, including Vidicon [trademark] , Saticon [trademark] , and
Plumbicon [trademark]. However, these names always have the same
implication: electron tube video pickup. A relative newcomer in the video
world is the CCD (Charge Coupled Device). Figure 1 contains an example
of a CCD camera. A CCD camera uses light sensitive semi-conductor
technology as a video pickup device. As mentioned later, CCDs have
gained a significant share of the market and will eventually totally displace
electron tubes as pickups. See figure 2 for an example of how a CCD pick-
up is installed in a camera.

There are several differences that can be directly attributed to the fact that
one technology is solid-state and the other is tube-based. First is sensitivity
to light. It is true for both CCD and tube cameras that more light helps the
camera generate better pictures. However, for low-level lighting, standard
CCD devices tend to produce a higher quality picture at a given light level
than their tube counterparts. Special formulations of CCDs are available to
make them even more sensitive in very-low-light level situations. For
very-high-level lighting situations, CCDs have the advantage, also. When
photographing bright lights, tube cameras tend to leave trails when
displaying the image, as the camera pans or the light source moves. This
image persistence can cause damage to the tube if the light source is
sufficiently bright or photographed for sufficient duration. CCD pickups,
on the other hand, are virtually indestructible when it comes to
photographing light sources. CCDs never have problems with image
persistence and can only be damaged by an intense heat source (e.g.,
created by focused light from direct photography the sun).

Another difference between solid-state and electron tube pickups is the
solid-state device has the ability to simulate a shutter with an adjustable
speed, while the electron tube device cannot. This is because of the
method in which each device obtains the image data. A pick-up tube must
continuously scan through a scene to provide a consistent image. For
example, if the tube camera was being used to capture a color image for
use by the television broadcast industry in the United States, it would
follow the National Television System Committee (NTSC) standard and
scan through the 525 lines of the image 30 times per second. A point of
the image would be read, and then the scanning beam would move to the
next point. This requires that the aperture between the scene and the
pickup device always be open. A CCD, on the other hand, obtains
information about the whole scene at the same time. Once the information
from the scene has been registered onto the CCD, the information is
translated to a scanning video signal by the electronics of the camera.
Because a CCD registers all scene information simultaneously, a shutter
can be placed between the scene and the pick-up. A high-speed shutter
provides the potential for crisp, still-frame images during action scenes
(like re-plays of sporting events). This is the same method that is used for
stop action in conventional photography. Like conventional photography,
however, use of a high-speed shutter demands more scene lighting to
obtain a good quality image because the imaging pickup device (film in
the conventional camera and the CCD in the video camera) is exposed to
the scene for a shorter period of time.

For low (240 lines, comparable to the VHS videotape format) or medium
(400 lines, comparable to Super VHS) resolution cameras, CCDs are much
less expensive to produce than low-resolution tubes. Because of this,
almost all cameras in this resolution range use CCDs as their pickup
device.

Until very recently, tubes were the only devices that could provide the
quality needed for high-resolution cameras (more than 500 lines). This,
however, began changing as high-resolution CCD cameras entered
production in 1998. As reliability and production of high-resolution CCDs
increase, CCDs will take over this segment of the market as they did for
low- and medium-resolution cameras. During the transition phase,
however, one must compare the advantages and pricing of both CCD and
tube cameras for high-resolution applications.

Another important innovation in video technology is the advent of the
digital camera. All of the traditional cameras mentioned above provide
output as an analog electrical signal that can be stored on tape or viewed
on a monitor. Recently, digital cameras have been introduced that output
the video as a stream of bits (binary ones and zeroes) that can be
understood by digital displays or digital recorders. One advantage of
digital video is the ability to make perfect copies because the bit pattern
used to create the displayed image can be replicated exactly. Another
advantage, specifically related to transmission, is digital video signals are
less prone to degradation over distance or in the presence of a weak or
noisy signal. However, once the signal crosses a certain signal-to-noise
threshold, the loss is usually total, not the gradual degradation experienced
in traditional analog systems.

3.2.2 Video Camera Features

The basic specifications of video equipment give the user a good, general
idea of how a unit should perform. (These specified parameters are
thoroughly explained in the next section of the guide, "Quality Parameters
and the User - Interpreting Manufacturers' Specifications"). Along with the
characteristics of a video camera that have a direct impact on its level of
performance, however, other features are sometimes offered that can either
give the user added flexibility and capability, or make the operation of the
camera easier under different conditions. In some cases, they can also
enhance the fundamental ability of the camera to perform better.

Below is a list of features that are offered in video cameras. Obviously,
they will not all be offered in all makes and models of cameras because
cameras are manufactured and sold to satisfy certain applications of the
perceived market. The more features, the higher the cost! If the
"importance" of these features is not currently understood, their merits will
become obvious later in the guide. Even without knowing the specific
details of individual video surveillance assignments for all readers of this
guide, it seems likely that most will want to consider a number of these
features. As a minimum, auto/manual white balance, auto-iris control, lens
compatibility, multiple mounting holes, and environmental robustness are
desirable features. As a starting point toward understanding the available
features, figure 3 shows the controls for the CCD camera shown in figure
1.

White balance and/or black balance: Automatic, manual or remote-
control-switch selectable. External white or black balance sensor possible.

RGB adjustments: Independent gain controls on red, green, blue, outputs.
Remote control of red, green, blue, and master gain (i.e., overall video
signal level).

Auto-iris control: For auto-iris lenses, an automatic gain control (AGC) or
variable gain (e.g., 6 dB) - selectable on/off.

Synchronization options: Internal (crystal) or external source.

Power options: Ac or dc power options, typically 12 V dc, 24 V dc or 115
V ac.

Lens Mount: Adjustable C-mount (adapter).

Lens compatibility: Accepts all (or most) types of manual and auto-iris TV
lenses.

Electronic shutter: Enables the camera to produce clear images in still or
slow-motion playback even when the objects are moving at very high
speeds.

Dynamic contrast control: Allows accommodation of scenes with a much
wider range of light levels than normal (e.g., allows detail in both sun-lit
and shadowy areas of the same scene).

Filters: Built-in optical filters can improve video results under various
lighting conditions (e.g., bright, subdued, inside, outside).

Viewfinder compatibility: A jack is provided to accommodate a
viewfinder, if desired.

Microphone holder: An adjustable ring or some other type of connection
apparatus is provided on the camera to mount a microphone.

Multiple mounting holes: Two or more tapped holes for mounting the
camera (e.g., on a tripod or wall-mounted bracket). More holes allow
different size (and weight) lenses to be accommodated while keeping the
assembly balanced.

Environmentally robust: Can operate in a wide temperature range (e.g., 14
[degrees] F to 122 [degrees] F) and can be stored in a wider range (e.g., -
22 [degrees] F to 158 [degrees] F). Can operate at altitude (e.g., 10,000 ft)
and in heavy relative humidity (e.g., 95 percent). Can tolerate shock and
vibration.

Another thing to be aware of is a camera does not usually come as an all-
inclusive video package. Components must be purchased along with it to
permit it to function. A few of the most common items required for the
camera system, but not supplied with it are listed below:

1. Ac power pack or ac/dc power supply

2. Lens

3. Coaxial cables (e.g., RG-59/U) for connections to recorder or monitor
from "video out" jack, from camera to external sensor or synchronization,
for remote control, etc.

4. Television monitor for focusing and other adjustments (such as white
balance)

3.2.3 Camera Lenses

One important component of the video camera system is the lens. A lens
for a video camera plays the same role as a lens for a 35 mm single lens
reflex (film) camera. It allows the user to capture an image in the camera.
Why all the different lenses? The difference in lenses is dictated by the
difference in shooting environments and the kind of pictures that are
needed. In a nutshell, the size of the subject, the distance to the subject,
and how much light is on the subject determines the best lens. Figure 4
shows a fairly simple, manual focus, manual aperture lens.

The primary specifications of all camera lenses are their focal lengths and
their f-stop ratings. Focal length is the distance from the center of the lens
to the point at which parallel rays from a distant subject come to a
common focal point. The f-stop number is the ratio of the focal length to
the diameter of the lens. These terms are explained below.

The size of an image that a lens forms inside the camera is determined by
three things - the physical size of the subject, the distance from the lens
(camera) to the subject, and the focal length of the lens. Lenses with a
short focal length (for example, 8 mm to 20 mm) are normally used for
wide-angle pictures and are called "wide-angle" lens. Lenses with long
focal lengths (80 mm to 300+ mm) are used to capture distant subjects and
make them look close. These lenses are generally called "telephoto"
lenses. A middle-of-the-line kind of focal length is 50 mm. Many new
single reflex cameras come with a 50 mm lens because camera
manufacturers feel it will give good overall performance for general
recreational uses on the average. It cannot do wide angle or telephoto
shots. Because lenses have different focal lengths, they also have different
coverage ratios. That means that their fields of view vary. The camera/lens
field of view is how much of the subject and the immediate surroundings
will be filmed.

A wide-angle lens has a tremendous field of view. If observing a person on
the street with a 16 mm camera system mounted on the side of a building,
several feet of the street, both sidewalks, and other buildings may be in
view. For a telephoto lens, just the opposite happens. A 300 mm lens
camera may capture the head of a subject at 100 ft and nothing else. The
light and viewing angles from the lens to the subject are very narrow. In
both simple cases given here, the ability of the lens to satisfactorily
capture the video image is dependent upon the light present. That fact
leads to a discussion of the f-stop parameter.

The light-gathering ability of a camera is determined by the diameter of
the lens. The larger the diameter of the lens, the greater the amount of light
falling on the subject. Lenses are rated at maximum diameter (i.e., the
largest iris opening). (The iris is either fixed at one size or is an adjustable
diaphragm that varies the opening for light to enter the lens. The opening
itself is called the aperture.) A lens' so-called f-rating is defined as the
focal length of the lens divided by its diameter (with the iris fully open).
The smaller the f-rating, the more light the lens can take in. This means
that a low f-number is needed when a scene has low light. High f-numbers
operate well in bright sunlight.

If a lens with a fixed iris and a 50 mm focal length had a diameter of 35.7
mm, its f-rating would be 1.4. The iris ring would probably have the rating
f-1.4 written on it. This lens would work well in a fixed location with
fairly low light. If a lens with an adjustable iris had a focal length of 25
mm and a diameter of 13 mm with the iris completely open, its f-rating
would be 1.9. On the movable iris, calibrated marks called f-stops would
indicate to the user that besides the f/1.9 setting, other settings for brighter
light conditions were also available by turning the iris ring. These f-stops
might be: 2.8, 4, 5.6, 8, 11, 16, and 22. The f-stop numbers increase in
steps so that each higher stop allows one half the light input of the
previous stop. For bright sunlight, f-22 is selected. Figure 5 shows a
camera lens with the aperture wide open and also with the aperture
partially closed. Auto-iris lenses control the light level automatically.

One important consideration in video camera operation that is affected by
the lens opening, f-stop, is the depth of field. The depth of field is the
distance between the object in focus closest to the camera and that object
farthest from the camera that remains in focus. An example of this is
clearly seen in a television scene when the camera is focused on a
performer close to the camera and the background goes out of focus. When
the f-stop is lowest (iris fully opened), the depth of field is poorest. To
capture everything in focus within the field of view, the camera system
must have the f-stop set as high as possible (ideally at f/16 or f/22). Once
lighting is reduced, however, only lower f-stops will capture any image
successfully. This correspondingly reduces the depth of field.

Zoom Lenses

Zoom lenses have continuously variable focal lengths. This reduces the
need to change lenses for different applications. Wide-angle and telephoto
tasks can be satisfied by the same camera with the same lens. A good
zoom lens may vary from less than a 10 mm focal length to more than 140
mm with an f-stop of only 1.8. Zoom lenses are specified by the ratio of
the minimum and maximum focal lengths. For a lens that can produce
focal lengths from 9.5 mm to 143 mm, its zoom range is 143 divided by
9.5, or 15. Product literature would express this ratio as 15:1 or 15X.

An important consideration in zoom lenses is whether or not the aperture
changes as the focal length changes. Variable aperture zoom lenses are
lenses whose maximum aperture changes, generally increasing, as the lens
is zoomed from smallest focal length to longest focal length. Fixed
aperture zooms maintain the same maximum aperture throughout the
zoom range of the lens. Fixed aperture zooms generally provide better
quality than variable aperture zooms. However, variable aperture zooms
are generally less expensive and smaller than their fixed aperture
counterparts.

Lenses for Special Camera Systems

A number of ultra-small color and black and white camera systems,
sometimes called microcameras, are available that have a separate control
unit and a separate camera head connected by a cable. (fig. 6.) The control
unit (box) contains the circuitry and adjustment controls (e.g., white
balance, noise reduction, high-speed shutters, internal/external sync)
normally found in and on the camera body, while the camera head
includes the lens and sensor subsystems. The reason for having such a
camera system is that it can essentially be placed virtually anywhere. It is
an ideal and versatile tool for several industrial applications such as the
observation of processes, material handling, quality control, and laboratory
experiments. In addition, it can be used very effectively for video
surveillance.

The camera control unit is small and light. Typical dimensions might be:
less than 4 1/2 in wide, less than 6 1/2 in deep, and about 1 1/2 in high. It
may weigh less than 2 lb. The cables, which connect the head to the
control unit, can be several lengths - from a few feet to almost 100 ft. The
camera head comes with a standard C-mount to accommodate a virtual
"catalog" of different lenses. Those lenses can be as diverse as the lenses
used in a normal one-piece camera system. They range from "super wide
angle" to telephoto, and many are available in a pinhole lens design.

Pinhole lenses vary in size but have one attribute in common: their front
element is very small and their overall construction mirrors this. Even with
lenses of focal lengths as low as 4.5 mm and as high as 200 mm, the
diameter of the lens barrel (the cylinder that houses the lens) is normally
only about an inch in diameter. The lens opening at the end of the lens
barrel may be just a few millimeters in diameter. (The conversion of
millimeters to inches is: 25.4 mm equals 1 in.) This means that a pinhole
lens could look through a hole (e.g., in a picture, wall or door), that was
about the size a pin would make. As an actual example, among the lenses
that Knox Security Engineering Corporation sells are two very different
products - the model SXZ4.5 auto-iris 4.5 mm lens and the model YX200
200 mm manual lens (fig. 7.). The wide-angle SXZ4.5 has a lens opening
of 1.6 mm, about 1/16 in. The "huge" 200 mm telephoto lens needs an
opening of 10 mm, a little more than 3/8 of an inch. Figure 7 shows the
profiles for these lenses and their barrels, and gives their specifications.

Another tiny camera system is called a "board camera." Board cameras are
CCD devices with a small lens and all other required electronics mounted
on an electronic circuit board. The board is typically 1.5 in square and
requires that video output cables and a power input source be wired (and
typically soldered) to the board. These cameras can be particularly useful
in cases where a traditional camera simply will not fit. Figure 8 illustrates
a board camera that is approximately 1.5 in square.

3.2.4 Still Video Overview

Still video, or digital still camera, technology was first demonstrated to the
public in 1981. The device demonstrated at that time was revolutionary. It
used a CCD to capture a video image, which was then stored on a
matchbook-sized floppy disk. The disk was capable of holding 25 frames
of video or 50 fields of video (a field is the equivalent of every other scan
line of a frame). The images stored on the disk could be viewed on any
standard television or printed on a special printer. The combined cost of
the disk and the paper and printer dye was significantly less than the cost
of a standard print. The problem with the device displayed in 1981 was a
lack of resolution. The still video camera at that time had a horizontal
resolution of about 200 lines, significantly less than a VHS recorder would
have. Because of the poor resolution available, actual production of still
video equipment was delayed until 1986. The equipment introduced in
1986 showed vastly improved resolution. The manufacturers had doubled
the initial resolution to something that is comparable with Super VHS, i.e.,
400 lines. Development has continued, and mainstream still video cameras
can now be purchased that have a resolution of more than 700 lines. There
are some very specialized digital cameras that have a resolution of more
than 2,000 lines. This is approaching the resolution of 35 mm film, which
has a resolution of about 4,000 lines to 6,000 lines, depending on the type
of film.

One of the most touted features of still video is the ability to take and view
pictures instantly. With still video, there is no need for messy, time
consuming and costly development of slides or prints. Also, the images
recorded on the disk can be sent in digital form anywhere instantly with no
loss of quality. Finally, unlike film, the floppy disks that are used to store
image data can be used again once unusable or unneeded pictures are
deleted.

3.2.5 Low-Light Cameras

Low-light CCD cameras can be categorized into two types: low-light and
low-light intensified. The simple low-light variety uses the same
technology as "normal light" CCD cameras and camcorders; the difference
is that their imaging chips are optimized for low-light conditions and/or
have additional refinements to be more noise-free over an expanded range
of incident light.

In low-light video surveillance situations, standard CCD chips have one
particular advantage over their predecessors, video pickup tubes: their
resistance to image burn-in caused by incident light that is too intense. If a
video tube were to focus on a scene that contained an overly intense light
source, an image of that light source would be permanently burned into the
tube and would be seen in all subsequent use, even after the scene had
changed. When devices are extra sensitive, as in the case of low-light
equipment, something as simple as a flashlight shined directly into the lens
can cause damage. Using a CCD camera in this type of situation helps
prevent damage.

CCD cameras designed for use in extreme low-light conditions
(approximately 0.05 lux and below[2]) usually have a device called an
intensifier built in front of the CCD array that multiplies the incoming
light for the CCD chips behind it. Modern intensifier technology exhibits
the same propensity to burn-in as video pickup tubes, however. Note that
in this case it is the intensifier and not the CCDs that exhibit this tendency.

The method of intensification enjoys widespread use because the use of
CCDs keeps the cost of the low-light camera below that of a video tube
type implementation, gives it ruggedness and durability, provides freedom
from frequent calibrations, and allows it to be used in a wider variety of
environments. If the intensifier were subjected to excessive light and had
some image burned permanently into it, the replacement cost for the
camera would be considerable.

Low-light intensified cameras can be found in a wide variety of price
ranges. Such devices range from attachments for existing cameras to full-
blown amplified-light surveillance systems, with the quality and
sensitivity of the device depending on the cost. Given they can be used in
rooms that would look completely dark to the human eye, a very sensitive
(therefore, very expensive) camera may be a small price to pay for law
enforcement or surveillance applications requiring this capability.
Regardless, the more typical application would be a poorly lit room or
night-time city scene, in which case a nominal sensitivity of 0.1 lux or
more would be sufficient for gathering evidence or performing general,
fixed, or mobile surveillance. Cameras claiming such sensitivities are
generally monochromatic (black and white), and do not employ the
intensification technology. They are significantly less costly than
intensified light cameras, making them a very attractive choice in many
applications.

Some practical examples are helpful to consider when deciding between
these two low-light camera technologies. A standard low-light camera
would probably enable a car's front license plate to be read at night, even
though the car's headlights were on. A low-light intensified camera
probably would not. In addition, the intensifier would probably be burned
and need to be replaced at a cost of around $4,000. If a very dimly lit
warehouse required surveillance, a low-light intensified camera involved
in that surveillance could be severely limited, if not altogether disabled, by
one very bright flashlight aimed at the camera's lens.

3.2.6 Infrared Cameras

Infrared cameras use special pickup devices that are sensitive to light with
wavelengths longer than those visible to humans. Within this category,
there are two types of equipment: thermal imaging systems and near-IR
systems. Both systems can be used in an environment that is totally dark to
human eyes but well illuminated from the camera's perspective. This
perspective changes, depending on the category of equipment.

Thermal imaging equipment is commonly used by the military for night
action. The image it forms is based on heat emissions from the subject it is
pointed at. The higher the temperature of the subject, the brighter the
image. This requires no special illumination but does require that your
subject be a different temperature from the background. Thermal imaging
systems are fairly expensive, ranging from $5,000 to $40,000.

Near-IR cameras use a special light source to illuminate the subject area.
While subjects may be in total darkness, the special light source makes the
scene appear bright-as-day to the camera. This special light source, an
infrared light, often will be provided with an infrared camera but can also
be purchased separately. Near-IR cameras with an accompanying light
source range in price from $700 to $1,500.

Some caution must be used in the selection of an infrared light source. As
wavelengths approach the red end of the visible spectrum (700 nm), it
might be possible for some humans to perceive the emitted light.
Therefore, it is desirable to have an infrared source with a wavelength of
more than 800 nm.

3.3 Camcorders and Recorder/Players

3.3.1 Video Tape Technology

Of the many video formats, VHS is the most popular in the world today.
Since the format's introduction in 1975, the popularity of the VHS system
has grown such that more than two-thirds of the households in this country
contain at least one piece of VHS equipment. The VHS format's primary
advantage is it is the lowest cost option for video. However, being the least
expensive format has its trade-offs: at 240 lines, it has the lowest
horizontal resolution of the available formats. (Resolution and other
important performance parameters are explained in section 4 of this guide,
called "Quality Parameters and the User - Interpreting Manufacturers'
Specifications.") The VHS format has also produced some variants, which
are on the market today. Included in these are VHS-C, Super-VHS
(S-VHS), and Super-VHS-C (S-VHS-C).

The "-C" designation implies that the system is compact. To achieve this,
the system uses a smaller cassette. The -C format is used almost
exclusively in camcorders because size and weight have a significant
impact on the camcorder user. Camcorders have been produced with this
format that weigh less than 2 lb. The smaller cassette employed by -C
systems still use the same size and type of VHS tape, but the smaller
cassette only holds approximately one-sixth of the amount of tape that a
"normal" VHS cassette holds. The amount of information recordable on a
tape is reduced accordingly. The small cassettes are playable in a standard
VHS machine with an adapter.

The "Super" designation indicates the same tape size and cassette as VHS
format but uses newer recording technologies to dramatically improve
picture quality. Super VHS equipment have a greater signal-to-noise ratio
and a higher resolution (400 lines) than the plain (240 lines) VHS. Tapes
recorded in standard VHS format are playable on Super VHS machines,
but Super VHS tapes recorded in S-VHS format are not playable on
standard VHS equipment. (S-VHS tapes recorded in VHS mode may be
played back in either VHS or S-VHS players with the 240-line VHS
resolution).

For several years after its introduction in 1974, the Beta format was
thought to be superior to VHS. As far as resolution was concerned, that
was true: Beta format has a resolution of about 260 lines. On other fronts,
however, Beta was not superior. Sony opted not to license the format to
other manufacturers, while licenses to produce VHS equipment were
readily available. The availability of a variety of equipment led to a greater
variety of prerecorded VHS material for public use. This drove the
popularity of VHS up while decreasing that of Beta. Eventually, market
forces led to an almost total stop-page of Beta equipment production. In
spite of the dearth of programming and the lack of mass market support,
there is still a small market for Beta equipment. This, however, will
continue to wane in the face of technology that is better and less
expensive. Beta did manage to produce one variation - ED-Beta (1985).
This format improved on the resolution of Beta, but failed to capture the
interest of the market. Its availability in the United States is very limited.

One technology that is currently gaining ground in the consumer
marketplace is 8 mm. This technology, introduced in 1985, uses a cassette
that is about the size of an audio cassette, yet will hold a full 2 h or 4 h of
video information. The resolution of this format is approximately 300
lines, somewhat better than that offered by VHS. Like VHS-C, the most
common use for this format is in camcorders. Also like VHS-C,
camcorders using this format have been produced weighing less than 2 lb.
A possible disadvantage of 8 mm, when compared to VHS-C, is the 8 mm
tape/cassette format is incompatible with any VHS equipment. However,
if the recording device, or another 8 mm camcorder or VCR system is
available, the tape will be playable through that device onto any NTSC
television or monitor. The tradeoff between available recording time and
compatibility has caused acceptance of the 8 mm format to grow slowly.
However, the installed base of 8 mm equipment appears to have reached
critical mass, as new home-based 8 mm equipment is becoming available
at prices only slightly higher than those available for VHS. The only
existing variation on this format is Hi-8, which has all the features of
standard 8 mm, but the resolution increases to approximately 400 lines.
Like Super VHS, tapes recorded in standard 8 mm can be viewed on Hi-8
machines, but tapes recorded in Hi-8 format cannot be viewed on standard
8 mm equipment.

A format introduced in 1982 that had sufficient quality for use in some
field production work is Betacam [trademark]. Betacam [trademark] was
phased out when a vast improvement was made on this system in 1986
with the introduction of Betacam [trademark] -SP. This change increased
the resolution from 320 lines to approximately 450 lines. This format is
currently very popular among circles where quality is very important, such
as television field production, and is used extensively in studios.
Panasonic has a proprietary format that is roughly equivalent to Betacam
[trademark] , called MII [trademark].

The final analog format under discussion is C. This format is for absolute
top-of-the-line NTSC analog video. It provides more than 600 lines of
resolution. It is the only major format that uses a reel-to-reel tape instead
of a cassette. This technology makes the equipment rather bulky, but the
size would be acceptable for use in surveillance vans. The bulk also
prohibits the use of the C format in camcorders and therefore requires the
use of a separate camera unit. There is, however, a price to be paid for the
quality of C format - the equipment is very expensive.

Recently, digital video recorders, cameras and camcorders have been
introduced. Professional equipment is currently divided into two camps:
Sony and Panasonic. Sony currently has five digital formats (D1, D2,
Digital Betacam [trademark] , Betacam [trademark] SX, and DVCAM
[trademark]) and Panasonic has three (D3, D5, and DVCPRO
[trademark]). Resolution of digital cameras is generally at least 400 lines.
Figures 9 and 10 illustrate a typical digital camcorder and some of the
controls one might expect to encounter.

Table 5 offers a list of video formats available for video surveillance
equipment, the resolution associated with those formats, and the price
ranges for equipment represented under each of the formats. The price
ranges include the prices for individual pieces of video equipment, that is,
for camcorders and recorder/players, unless otherwise noted. (For instance,
for the VHS-C format, there are no VHS-C player/recorders on the market.
VHS-C cassettes are played in normal VHS player/recorders with an
adapter. The figures in the price range reflect the prices of VHS-C
camcorders only).

3.3.2 Camcorders and Video Recorder/Players Features

Camcorder and video recorder/player products offer a vast number of
features. Many of these are well known, while others are not obvious, and
are therefore not considered very often. Some of these "subtle" features
may be just what are needed for certain kinds of surveillance applications.
A number of the commonly advertised features are briefly explained
below, along with some of those receiving less attention.

Auto/manual focus: Automatic focus will change the focus based on the
perceived target and maintain it until something changes. Even if auto
focus is available, professionals often will use manual focus in cases when
there is a chance that automatic feature will have trouble differentiating
the target from other activity.

Auto/manual white balance: Automatic white balance will maintain the
optimum color balance in either indoor or outdoor conditions. Manual
control is useful if unique conditions exist that the auto white balance
feature cannot deal with (e.g., strong backlight).

Auxiliary microphones and jacks for special applications: Many
camcorders will accommodate the use of auxiliary microphones if the
built-in microphones are not adequate for special applications. A desirable
characteristic of some camcorders is the existence of a separate jack for
the auxiliary microphone. With the jack, it is not necessary to remove the
basic microphone from the camcorder. This keeps that microphone stored
on the camcorder and eliminates the chance of it being misplaced.
Wireless microphone sets, like the Vivitar WMK-2 Wireless Mike Outfit,
transmit over radio frequencies for hundreds of feet and are capable of
transferring clear audio through typical doors and walls.

Battery type: The ability to use batteries allows portability. Equipment
might use rechargeable batteries (e.g., nickel-cadmium or lithium-ion) or
single-use batteries (e.g., alkaline). See NIJ Guide 200-98 for more
information on batteries.[3]

Book mark search: With this feature, one may return to the point where
recording had previously ended.

Day/time setting: A built-in calendar and clock allows each recording to
be "stamped." A "button cell" battery keeps the date and time correct.

DV in/out jacks: Many digital camcorders have a special jack for digital
video input and output. These input and output signals are most frequently
based on IEEE Standard 1394, also known as "FireWire." Personal
computer interface kits can be purchased from both Sony and Canon that
will allow digital video to be downloaded from one of these camcorders to
a PC.

Edit controller interface: LANC is the most widely available interface for
camcorders.

Fade control: When this control is activated, the picture in the viewfinder
of a camcorder and on the tape will fade out. When the control is
disengaged, the picture will automatically fade back in.

Flying erase head: Allows user to make exceptionally clean edits of the
video tape. Video and audio "dubbing" (i.e., changing) is possible.

Headphone jack: Usually a 1/8 in stereo phono jack.

High speed shutter switch: Allows the camcorder to capture and record
high speed activities for slow motion or still playback. Speeds might
include 1/250 s, 1/500 s and 1/1000 s.

Image stabilization: Optical or electrical.

Index search: An index mark can be placed at the beginning of each
recording so that automatic review and playback can be accomplished
easily.

LCD monitor: Provides viewfinder information in a larger screen so
camcorder does not have to be held to the eye. Typical sizes range from 2
in to 4 in, color or black and white. Figure 11 illustrates a camcorder with
both LCD monitor and viewfinder.

Light source: Built- in or accessory.

Macro control: This is used to unlock the zoom lens on a camcorder so
that it can be used to get in-focus close-ups of subjects normally too close
to shoot.

Motion sensor: Useful for situations requiring constant surveillance but
where a low activity rate does not justify constant recording. Motion
sensor activates recording function. Audio sensors are also available.

Multiple heads for still frames/slow motion playback: Video head design
is an area where significant improvements have been made in the past 5
years. In simple systems, one video head is required to record and
playback the video track. Multiple heads have been added to improve
quality at different speeds, with some units automatically switching the
output from head to head to maximize the amount of signal that can be
recovered from a tape. As a result, noise-free still frames and slow motion
effects can be produced. It is also possible to have less noise (snow) in the
picture scan mode.

Multiple start/stop buttons for diverse operating conditions: Some
camcorders have more than one "start/stop record" button - the Ricoh
R800 for instance. One button is found in or near the grip and the zoom
switch. The other button is located on or near the lens barrel, for use when
the camera is in a low position and a normal grip is impractical.

Noise reduction: Improves the picture quality in marginal lighting
situations.

Optical and digital zoom: Many camcorders, especially digital camcorders,
provide a combination of optical zoom and digital zoom. Optical zoom is
the traditional method of providing zoom lens capability. It is achieved
through rearranging the distances between some of the optical elements of
the lens. As the magnification of the subject increases, generally the
camcorder loses some of its light sensitivity. Digital zoom is achieved by
expanding a small area of the image pickup device (specifically a CCD) to
fill the whole screen. This generally results in a loss of image sharpness
and/or resolution.

Photo mode: Usually available on digital camcorders, less frequently on
analog models. This feature allows a single frame of video to be recorded
across several seconds of tape.

Remote controls: Typically using wireless technology, remote controllers
can start and stop recording and even control zoom functions.

Self timer/time-lapse recording: Self-timers are an attractive feature for
surveillance that involves predictable patterns. Camcorders and recorders
can be programmed to start and end at a certain time, or to record only 1 s
each minute for several hours.

Sensitivity/gain-up controls for shadows: "Sensitivity/gain-up" controls or
buttons are intended to increase the brightness in scenes that need it. The
better sensitivity mechanisms have made shadowed images 50 percent to
80 percent brighter with little increase in noise or distortion (graininess in
the image).

Special effects (Special FX): Examples include fades, wipes, solarization
and posterization. These functions produce interesting visual effects but
are probably not very useful for surveillance applications.

Tape and time counter: Displays a number reading for the position on the
tape or the elapsed time during a recording. Time remaining may also be
displayed.

Wireless playback: Using wireless technology (e.g., infrared
transmission), camcorder recordings are played back using a display that
has an appropriate receiver (typically supplied) for the wireless
transmission.

Titling: Annotations of various lengths and types can be added to the tape.

3.3.3 Camcorder Accessories

In addition to the features that come directly with a camcorder, other
accessories can be purchased that make the tedious and difficult
operational tasks of the camcorder a little more tolerable. A few are listed
below. Others are available where video equipment is sold.

Supports                  Order your surveillance system

Even with a small camcorder, it is a chore to hold it in front of you for any
period of time. Fortunately, certain devices are available to help you carry
your camcorder. One such device is called SteadyCam [trademark] , and
another is called Glidecam [trademark]. These units have a harness that
attaches the camera and steadying mechanism to the body and an arm that
holds the camera out in front of you and swings to the side. The user has
both hands free to manipulate the camera or camcorder. The price of such
an equipment aid varies from about $200 to about $4,000, which may be
well justified if several hours of "hand-held" video taping is needed.

Another type of support is a basic shoulder mount. This kind of item
comes with a padded shoulder rest and handle and costs about $200. The
tripod mount of the camcorder is attached to the product which allows the
camcorder to be adjusted back and forth to suit the user's eye position.
Some may find this device awkward to use and somewhat less comfortable
than a harness system, especially if the camcorder is long or front-heavy.
An example of a shoulder mount product is the Videosmith's
MightyWonderCam [trademark].

The conventional tripod is yet one more type of support product.
Unfortunately, many of the sturdy tripods tend to be bulky and awkward to
carry. A number of products on the market, however, including the
Cullman Video Magic Tripod, use aluminum materials and good folding
designs so that units weigh less than 3 lb and can be collapsed into a very
manageable 14 in x 6 in x 1« in shape. The Cullman product also includes
a built-in monopod, a two-way pan head with handle that allows easy and
fluid panning. The Cullman unit has a manufacturer's suggested retail
price of about $200.

Auxiliary Monitors

Most camcorder viewfinders are quite small in size (about an inch square)
but provide the user with a relatively good image of the scene in view. At
the same time, they can provide much data about the current operation of
the camcorder (e.g., time remaining on the tape, manual or auto settings,
and battery condition). With all of those data appearing and the necessity
to concentrate on the scene to be recorded, some users may prefer to deal
with a larger display. Some camcorders come with larger displays in
addition to or instead of the traditional viewfinder. For those camcorders
that do not have a larger display built-in, auxiliary monitors, which fit on
the accessory "hot shoe" found on many camcorders, offer one way to see
a larger color image of the scene. They also allow the user to review
camera settings and status without constantly glancing down into the
viewfinder. One such monitor is the Citizen LCD Color Monitor. It
weighs only 6.3 oz, has a 3 in diagonal screen, and can be powered by AA
batteries (for 3« h), battery pack, or AC.

Environmental Enclosures

To expand the utility of camcorders, many companies offer environmental
enclosures for cameras and camcorders. These range from simple rain
covers to underwater enclosures. Rain covers vary in price from $50 to
several hundred dollars, depending on manufacturer, degree of protection,
and features (some come with heaters!). Underwater housings vary in price
from $600 to several thousand dollars, depending on manufacturer, depth
rating, and camcorder model.

Video Capture Cards

With the advent of "multimedia computing," a large number of computer
interface devices have become available that allow video to be imported
into a computer for inclusion in reports, presentations, and to be printed on
computer printers. The devices are varied and command a wide range of
prices. At the bottom end is the Snappy Video Snapshot, a device that
hooks to a computer's printer port and converts a composite analog video
frame to a computer image. More elaborate systems include cards that
plug into the expansion slots inside the computer. These cards might have
video input/output jacks and record full motion video to the computer's
hard drive (e.g., Data Translation Broadway Beginner).

For digital camcorders with a DV interface, there are special kits available
from the manufacturers of the digital camcorders to allow easy download
of the digital video from the camcorder to a personal computer. One
example is the Canon Video DK-1 DV Capture Kit. The kit includes an
IEEE 1394 interface card for the computer; a cable to connect the digital
camcorder to the computer; and software to control the digital camcorder,
download images, and save the images to disk.

3.3.4 Format Applicability to Surveillance Requirements

Following are several tables that provide information on what type of
equipment would be applicable to specific surveillance conditions. The
equipment recommendations conveyed in the tables are oriented toward
the need to perform real-time information collection. Equipment not
meeting these requirements are not shown in the tables. For example:
while a monitor/television is required to display information on a
videotape, it is not necessary to be able to view the tape in great detail at a
crime scene. However, for some surveillance applications, such as where a
high power or amplified light lens is being used on a camera, a monitor
may be necessary to make sure the subject is being properly recorded for
later presentation (e.g., in court).

Table 6 provides a list of video equipment applications along with
recommendations on what equipment types would produce meaningful
video data. Table 7 provides a bit more focused detail by recommending
equipment based on specific surveillance performance requirements.

A note of clarification regarding table 7 is in order since standard VHS
and 8 mm equipment was generally not recommended because of its lower
quality (i.e., resolution). Higher-resolution Super VHS and Hi-8
equipment is now widely available at reasonable prices. If, however, cost
is the most important factor in the decision of what equipment to purchase,
the standard VHS and 8 mm formats should produce moderate results for
any performance requirement where Super VHS and Hi-8 is specified.

Another consideration when determining what equipment will meet a
performance requirement is light level. In all cases, camera resolution
drops when light drops. The lowest light level in which manufacturers
claim their cameras can acceptably record a scene varies from as high as
30 lux to less than 1 lux. Thirty lux corresponds to the lighting expected in
an underground parking garage, and 1 lux would be equivalent to a
medium sized dining room lit by two or three candles. It is the 1 lux rating
of some of the camcorders available on the market that allowed them to be
included as a possibility for meeting performance requirements in dim
lighting conditions.

3.4 Monitors/Televisions

3.4.1 Technology Summary

Since computers and video equipment both use monitors, one might
conclude that a computer monitor should work with a video camera
system, right? However, this is not the case. Computer displays have a
different function than the video displays used for industrial or broadcast
video. The video systems used throughout the world for such purposes as
recreation, news, education, and surveillance were designed and
implemented to broadcast moving pictures. Motion tends to make humans
focus their visual attention (and their need for sharpness) in the center of
the screen, with the corners and edges of the screen treated as only a
secondary concern. Video systems display objects that people recognize in
real life. They also count on people to "help the images along" by using
their experiences to fill in lacking details and color as required. This is
often needed because video systems that people are most frequently
exposed to, such as television, are medium-resolution black and white
systems with a low-resolution color channel overlaid on the black and
white image. Even the "high-definition" television (HDTV) systems
proposed as the next generation of TV (that will have about twice the
resolution of existing systems) will continue this approach.

Computer systems display static images of detailed information, such as
words or numbers. Picture elements (pixels), which can be thought of as
little dots of light, are grouped in patterns to form the lines, letters, words,
numbers, and other symbols seen on a computer monitor. Information
found in the corners and on the sides of a computer monitor is just as
important as that found in the center of the screen. The graphics created by
computers and their displays, although much improved from just a few
years ago, still do not perfectly reflect natural things but tend to be
abstract.

The difference in applications between computer monitors and video
monitors will help explain the rationale behind the basic technical design
of video monitors/TVs and why computer monitors would not work very
well for video applications (even if the interfaces were compatible). Note
that video monitors and televisions are considered to be essentially the
same in this discussion, since the way video is formed, transferred, and
received is fundamentally the same for both. To be more explicit, the
monitor is like a TV receiver, with the picture tube and associated circuits
but without the rf (radio frequency) tuner and if (intermediate frequency)
section. A true monitor does not have antenna input connections to receive
radio broadcasts but receives video from other sources via video input
jacks. Several products are on the market that are combination
monitor/televisions. Dual sets of connectors allow either TV or monitor
applications.

The ability of a video monitor/TV to resolve an image, that is, to show the
detail of the image, is limited by two bandwidth restrictions -
approximately 4.7 MHz for the luminance (black and white) and 1.5 MHz
for the chrominance (color) portions of the picture. These bandwidth
figures apply to the RS-170A video standard used in the United States to
define NTSC video (color television). The specifications were chosen to
conserve radio frequency spectrum for the broadcast services, and because
of the limitations of television technology at that time. These bandwidth
limitations still apply regardless of how well the equipment is designed
and built.

(Computer monitors, by the way, have a much wider bandwidth, typically
from 20 MHz to 100 MHz. No real standards restrict the design of
computer monitors; only technological and economic factors apply. With
virtually an unlimited amount of bandwidth available, computer displays
can show a tremendous amount of detail.)

For monitors, usually color is fed through 3 separate signals - red, green,
and blue - with identical bandwidths. (The red, green, blue signals are
where the acronym "RGB" comes from in video literature.) Picture tubes
used in computer displays have a much smaller, highly-focused electron
beam spot size and finer pitch screen surface than most video
monitors/televisions.

Besides video bandwidth, another technical parameter that affects the
"\definition, or the quality of detail on a display screen, is scan rate. The
video picture is scanned in a sequential series of horizontal lines, one
under the other, to permit one video signal to include all the elements for
the entire picture. In effect, video pictures are reassembled line after line
and frame after frame. For NTSC video, a total of 525 lines are required
for the development of one picture (frame). All 525 lines are scanned in
1/30 of a second.

The higher the horizontal scan rate and video bandwidth, the higher the
resolution. In addition, for a given horizontal scan rate, as the vertical scan
rate decreases the level of detail increases because there are more
horizontal lines used to make a complete image. Like the bandwidth, scan
rates for NTSC video are specified in much detail in the RS-170A
standard. The broadcast standard mandates a horizontal scan frequency
(rate) of 15,734.263 Hz. This number is commonly referred to as 15.75
kHz. The vertical scan rate is fixed at 59.94 Hz and is normally called 60
Hz. For comparison purposes, computer monitors have horizontal scan
rates between 15 kHz and 100 kHz. (Once again, no standards restrict the
rate.) At 75 kHz, the computer monitor is almost five times faster than the
video monitor.

Vertical scan rates for computer monitors run from about 40 Hz to 120 Hz,
but many of the video cards available for computers today start with a
default vertical scan rate at or around 60 Hz. This is a compromise
between having the lowest possible vertical scan rate and having problems
with people viewing the screen. Vertical scan rates below 60 Hz are
somewhat of a problem for humans. If the scan rate is not fast enough to
prevent the light emission from the phosphors in the display from
decaying too far, the resulting variations in the brightness of the image can
be seen. The varying brightness is perceived as a definite flicker.

With all of the constraints placed on video, it is no wonder that some
people have compared the best resolution possible for a video monitor and
a computer monitor to the difference in picture quality between a
newspaper and a magazine, respectively. Nevertheless, beyond its image
detail (resolution), a few more picture quality measures can be used to
describe a good monitor's performance. Assuming it is synchronized to
stay still, a color or monochrome (black and white) monitor's reproduced
picture should also have high brightness, strong contrast, and the correct
proportions of height and width (aspect ratio). Also, color sets should have
strong color, or "saturation," with the correct tints or hues.

3.4.2 Monitor/Television Features

As mentioned above, many of the characteristics of video monitors and
televisions are fixed by a recognized NTSC standard so that video
broadcasts may be received equally well by all. Even so, monitors will be
offered with various ratings for quality parameters as basic as resolution.
Resolution for monitors/televisions will range from 200 lines to 300
lines[4] for inexpensive models found in the home to units with 400 lines
or 500 lines for those with discriminating (and expensive) taste. Units with
800 lines to 1,000 lines are used in television broadcast studios. One way
to gauge the resolution needed for a particular application is to be aware
that the best resolution one can expect from over-the-air broadcast or cable
TV service is 330 lines. If a "good" TV picture will suffice for a certain
task using a monitor, it is not necessary to select one with more than 330
lines.

Other featured items to be aware of when contemplating monitors include:

Screen size: Measured diagonally, this can vary dramatically. Typical
sizes run from about 8.5 in through 20 in, but super-small and huge
monitors are available, also.

Color or black and white presentation: Both are available at many
resolution ratings.

Built-in speaker, jack for external speaker, headphone jack: Allows audio
monitoring publicly or privately.

Selectable inputs: BNC-type coaxial cable and/or 8-pin video jacks for
composite and RGB video are available. Switchable line, camera, and
VCR input jacks may also be offered.

Monitor bridging: Selectors and connectors allow bridging to display
video on multiple monitors simultaneously.

Synchronization signal: External input and output synchronization
interfaces for when synchronization with a separate video device is
required.

Front panel controls: Include brightness, contrast, vertical hold, horizontal
hold, tint, and color.

Blue-only control: This displays only the blue electron beam for simplified
adjustment of chrominance and hue using a color bar signal.

Comb filter: Integral to a monitor's design, a comb filter minimizes loss of
resolution and reduces streaking and wavy edges on fine patterns.

Remote control: Wireless.

Input power: 120 VAC and DC versions are available.

Mounting options: Rack mountable.

Carrying handle: Folds down when not required.

Enclosure: Metal cabinet and magnetic shield ring reduces interference
from other electronic equipment.

3.5 Special Surveillance Systems

3.5.1 Specialized Camera Systems

In recent years, the electronics industry has revolutionized the video
camera industry. Use of CCDs and integrated circuits have allowed a
considerable reduction in the size and cost of video cameras. One product
that is available is called a "board camera." This camera consists of the
CCD and other electronics on a 1.5 in square (or smaller) printed circuit
board with a lens mounted over the CCD. Because of their small size, they
are easily concealed. Some examples of places these cameras can be
concealed include ties, hats, jacket lapels, brooches, books, cigarette
packs, smoke detectors and briefcases. Power is supplied by an external
device such as a transformer or battery pack. The video signal is usually
fed to a monitor, video recorder, or video transmitter.

3.5.2 Patrol Car Surveillance Systems

Patrol car surveillance systems are special video (and audio) equipment
ensembles that were designed specially for police applications. Originally
conceived to be that silent partner for individual officers on patrol, the
applications for these systems have expanded beyond officer safety. Not
only do these systems provide a clear record of faces, vehicles, license
numbers, weapons, and the conversations that transpired before and during
dangerous situations (so that back-up can be called in), but videotape
documentation of routine occasions has also been found to be valuable.
Videotapes have been critical evidence in allegations and liability suits
against police and have been used extensively in contested arrests,
particularly drunk driving cases. In addition, video and audio tapes from
patrol car surveillance systems can be used as training tools for new
officers (or experienced officers) to insure proper procedure and caution
are exercised under various circumstances.

A typical patrol car surveillance system consists of a camera, control and
status panel, recorder (either 8 mm or VHS tape format), protective case
for the recorder, and wireless microphone. The camera is mounted on the
inside of the police car's front windshield. It is a digital CCD black and
white or color camera that normally can operate across a wide illumination
range (from low light provided by headlights to bright sunlight). An auto
iris lens adjusts the light level from day to night viewing, while a
polarizing filter is used to reduce reflected glare. Since the camera has
been designed for the police application, it is small, lightweight, and
resistant to vibration and shock. A wide-angle lens (e.g., 8.5 mm to 15
mm) allows the camera to view an extended area.

The control and status panel is located near the officer in the car. It may be
installed next to the radio, for instance. This unit allows the officer to turn
the system on manually or to have the system come on automatically when
the overhead flashing lights are turned on. The recorder stops when the
officer selects the off control. The unit also indicates the status of the
recorder and the microphone. A display light or other type of warning is
given when the recording time is nearing or at its end. A VHS tape records
up to 6 h of video and sound; a 8 mm tape can record 2 h.

The recorder, in an environmentally controlled, fireproof, bullet-resistant
case, is usually located in the trunk of the vehicle. Heat or cooling is
provided into the case when thermal switches detect a need. Limited
access to the trunk and into the recorder case (it can be padlocked) helps
protect the tape from tampering and preserves its integrity as evidence in
court. The recorder itself cannot be removed (even for playing back tapes).

A tiny wireless microphone, used in conjunction with a pocket-sized
transmitter and antenna, allows the surveillance system to hear sounds
around the officer, especially when he/she leaves the patrol car. The
microphone can be attached to a lapel or tie and may be provided with a
wind-screen to greatly reduce background noise caused by the wind. The
transmitter and built-in antenna can be clipped to a belt or be kept in a
pocket. The wireless microphone has a range of about 1,000 ft (officer to
car) under normal conditions. Because its range is limited, a radio license
is not required for this transmitting system.

Another component of the police car surveillance system that may be
offered is a video/audio monitor in the car that can be used for continuous
viewing and for focusing and adjusting the camera. If it is not practical, or
too expensive, to install a monitor in each patrol car, one monitor may be
used to focus and adjust the cameras of several (or all) surveillance
systems in a department.

An example of a patrol car surveillance system is the Eyewitness
[trademark] system sold by Kustom Signals, Inc. of Lenexa, Kansas. A
complete system is priced between $3,900 and $5,500, depending on the
type of video tape format required (VHS or 8 mm). The Eyewitness
[trademark] system includes either a color camera that has a minimum
illumination of 5 lux and 300 lines of horizontal resolution, or a black and
white camera that can operate at 0.5 lux and 420 lines of resolution. Both
cameras can operate from 14 [degrees] F to 122 [degrees] F. The selected
camera is connected to either an 8 mm or VHS video recorder that resides
in a patented "vault" in the trunk.

3.5.3 Retractable Surveillance Systems

Designed to replace conventional overhead closed circuit television
systems, these specialized video surveillance systems take many shapes
and sizes. Some, such as the Knox Forward Intelligence Gathering System
(FIGS), are in-ground/above-ground products designed for both industrial
and government applications. The FIGS camera head assembly can be
buried in the ground, hung from a pole or traffic light, or fitted into the
recesses of a building. When activated, the camera head emerges beyond
the edge of its case to a desired height at or below 8 in. In its basic
configuration, FIGS will connect to most pan-tilt control drivers for full
control over its main functions (including vertical and horizontal viewing,
focusing, zoom, iris adjustment, and other auxiliary needs). Various
cameras can be used with FIGS - a standard black and white CCD;
optional high resolution B&W CCD or color CCD camera, or optional
intensified day/night camera.

The control unit for FIGS is available in a waterproof, air-tight carrying
case, and can control the camera assembly unit via wire, or optional UHF
radio control link. Video information may be transported by wire or an
optional microwave radio video link. For law enforcement and military
applications, FIGS can be obtained with a host of special electronics. For
inner city surveillance, both data and video can be transmitted over
dedicated phone lines using special line drivers. FIGS can also operate
over satellite.

Another Knox product that is similar to the FIGS, but fits well into another
environment, is the Covert Car Antenna Video System. Details on this
system, and others, may be acquired through the manufacturer in
Greenwich, Connecticut, or other makers of video surveillance gear.

3.5.4 Portable Systems

If it is not possible to monitor an area from afar or if subjects frequently
move from one location to another, it may be necessary to go to where the
information is. For just those occasions, undercover attach‚ cases are
available. One such case is made by ESC. The internal components, which
consolidate the image and audio capturing and transmission functions, are
cleverly hidden in a false top compartment of the case leaving no visual
clue as to their existence. The tiny hole in the case, through which the
camera operates cannot be seen even from as close as a foot away.

A 9 mm f/3.5 pinhole wide-angle lens is interconnected to a CCD camera
that has a minimum illumination rating of 3 lux, resolution of 280 lines,
and a signal to noise ratio of 46 dB. A number of video link options are
offered that include UHF and microwave radio transmitters and matching
receivers. The 1.3 GHz system also comes with a mini-dish antenna. All
systems are powered by batteries that fit in the case.

--------------------------------

4. Quality Parameters and the User -- Interpreting Manufacturers'
Specifications

The most relevant technical parameters used to measure the quality of
cameras, camcorders, video recorders/players, and video monitors are
described in this section. There are very specific relationships between
these parameters that engineers use to assess the quality of video gear and
what the typical user notices when using the equipment. The explanations
of the performance parameters, therefore, contain "real world" information
related to the human perception process along with the basic definitions of
the engineering terms. Once a user understands how a particular technical
parameter will affect him during video surveillance work, he can relate
data from the manufacturers' brochures and specification sheets to his
needs. Equipment selection and purchase then becomes easy.

4.1 Technical Parameters' Relationship to Law Enforcement and
Corrections Needs

In the earlier section on video surveillance requirements, it was suggested
that a large number of specific video needs for law enforcement and
corrections could be summarized into a short set of general requirements
(e.g., identifying subjects, recording data) Table 8 shows how the
numerous technical parameters that are used to describe the functional and
physical characteristics of equipment, and to measure its quality of
performance, correlate to this fundamental set of police needs. That is, the
table indicates what parameters might be especially important to consider
for certain applications. In several cases, one parameter (e.g., resolution)
can be seen as being relevant to more than one community need. (In fact, it
can be argued it should be listed under all need categories.) The table is
not intended to include all possible combinations under all conditions.
Rather, it is intended to be a "jumping off point" for users to contemplate
when they start to review what technical parameters are important to them.

The following section introduces the technical parameters and explains
their basic concepts. This background material should be helpful in tying
parameters to user applications and requirements.

4.2 Parameter Definitions

4.2.1 Resolution

The parameter most often quoted as being a reliable measure of quality is
resolution, which is the capability of a piece of video gear to distinguish,
record, and/or reproduce the details in a scene. The higher the number of
"lines," or "TVL" (Television Lines), the greater the horizontal resolution
of an image. A look at the method used to determine the number of lines is
helpful in better understanding resolution.

The number representing the measured resolution is arrived at by first
focusing a camera at a test pattern, which typically has alternating black
and white vertical lines of equal width. (A number of test patterns may be
present on a single chart.) By situating the camera a certain distance from
the chart, a pattern can be made to fill the entire view of the camera. If the
camera is connected to a monitor or TV, this is easy to do by simply
observing the monitor or TV. If the camera is close to the chart, it is not
too difficult to find the number of lines filling the view of the camera
simply by counting them on the monitor.

The number of lines seen by the camera can be increased by moving the
camera back away from the chart or by zooming back with the lens. If this
process is continued, there will come a point where the vertical lines are
too close together for the camera to "resolve" them, that is, to see two
neighboring white lines as being separated by a black one or two
neighboring black lines as being separated by a white one. Instead of
alternating black and white lines, the camera will begin to see a uniform,
medium gray. Before this happens, the number of vertical lines across the
screen is counted, and this number is noted as being the "resolution" of the
camera. Note that "240 lines of resolution" means that a device can
distinguish, record, and/or reproduce at least 240 lines of resolution.

For the above test to work as described, the monitor must be of higher
quality than the camera. If a monitor is being tested, the camera providing
its input must be of higher quality than the monitor. For a more precise
measurement of resolution, a high-quality digital oscilloscope with a
television synchronization option is used to view and measure the
electrical output of a camera. Lines can be counted automatically between
two cursors marking the edges of the test pattern's waveform.

Now that it is known that more resolution is better than less, how many
lines of resolution are really necessary in a piece of video equipment? The
answer depends upon the application. Some examples may provide a rule
of thumb. Three hundred thirty (330) lines of resolution is considered to be
the quality limit of what can be received by broadcast or cable television
in the home. Two hundred forty (240) lines is the nominal figure for VHS
or 8 mm video tape formats. People notice that VHS and 8 mm tape
playbacks do not resolve individual hairs on a person's head when the
person fills the screen to the extent that a newscaster does at a normal
distance (head and shoulders - referred to in the industry as a "talking
head"). If the newscaster is viewed from a broadcast ("over the air"), much
more detail may be visible, and individual hairs probably will be
noticeable. Super-VHS and Hi8, with common resolution figures of 400
lines to 420 lines, definitely are able to resolve hair detail at this distance.
Higher quality studio-type equipment, such as Betacam [trademark] -SP,
and proposed HDTV broadcast, cable, and tape standards may reveal even
the pores on a person's face at a "talking head" distance.

The bottom line for resolution is this: the higher the resolution rating, the
easier it is to positively identify a suspect who is across a parking lot or
street. Better resolution also means the videotape played back in court will
show more detail. What really is necessary though, since more resolution
means more money? The answer can follow this rule of thumb. If the live
image or recording has to be as good as the best broadcast or cable TV
picture you have ever seen, at least 330 lines of resolution are necessary.
(That means Super VHS, Hi8, digital, or broadcast quality in a
camcorder.) If you can live with less detail in most cases, select a VHS or
8 mm camcorder, or a camera, player/recorder, or monitor with fewer lines
of resolution.

4.2.2 Signal-to-Noise Ratio

Electrical and electromagnetic signals are all around us constantly. They
originate at both natural and man-made sources. Many of the signals
provide information for us and are desirable at a particular point in time
(e.g., we tune our receivers to find signals broadcast from commercial
television and radio stations). On the other hand, some signal emanations
interfere with, or detract from, the information in the desired signals that
we are trying to receive. These signals are called noise.[5] With any type
of telecommunications transfer involving audio (including voice), video,
data, or multimedia (i.e., combinations of audio, video, and/or data), it is
important to maximize the desired signal (or signals) and minimize the
noise to accurately receive the information. The signal-to-noise ratio,
SNR, is a measure of how well this has been achieved. The SNR is the
power of the desired signal divided by the power of the noise signal.

During video surveillance work, a number of information signals are
transferred. For example, a camcorder takes video and audio information
collected by its camera and microphone and records it onto a magnetic
tape in its recorder. Similarly, a stand-alone video camera passes visual
information to the tape of a separate video recorder or to a monitor for
viewing. If noise is present during any of the transfers, the quality level of
the video and audio information will be degraded. The tape will record
both the desired and undesired signals, and the monitor will display the
noise along with the video. Evidence of noise in video transmissions is the
appearance of "snow" on the screen, which essentially dilutes the video
signal. Audio noise is commonly heard as popping or hissing sounds.
When noise levels become too high (as compared to the desired signals),
the video image will be completely lost in a "whiteout" or in a wash of
distorted colors. The audio information will be indiscernible. In many
operational settings, the level of noise will not change very much over a
certain time span. Therefore, if the surveillance equipment on hand is not
adequate to properly capture and record the desired information under the
conditions present, there is little the operator can do.[6] It is critical, then,
that equipment be procured with a signal-to-noise ratio that is sufficient
for prospective operational scenarios.

The developers of video equipment realize the detrimental effects that
noise can have on video quality. Therefore, designs are employed that
inherently reduce the system's sensitivity to unwanted electrical noise. At
the time of manufacture, noise shielding may also be installed in areas of
the product that are still susceptible to noise degradations. Overall,
different manufacturers use many techniques and attain various levels of
success as they attempt to protect their equipment from noise.
Unfortunately, it is not clear to the casual observer examining equipment
which products have the greater natural immunity to noise. The SNR
specification, if provided, can be looked upon as an immunity indicator,
however.

The lower the noise sensitivity (the higher the SNR), the greater the ability
to get a quality image out of a device at low light levels. But how is SNR
quantified? The SNR is measured by putting a known high-quality signal
into a piece of video gear, recording what comes out, and comparing the
output to the input. Since some noise always gets into the signal by the
time it is output, measuring exactly how much noise was output can
provide an SNR.

For cameras and camcorders, most specification sheets will show SNR in
"dB" at a recommended illumination (light level). The abbreviation "dB"
means decibels. Decibels in this case do not have anything to do with the
magnitude of sounds or loudness. Decibels simply express the logarithmic
ratio of two voltages, for the signal and the noise. A ratio of 6 dB
(technically 6.02 dB) means the signal is twice the noise. For each
additional 6 dB, the voltage ratio doubles (e.g., 12 dB = signal 4 times
larger, 18 dB = signal 8 times larger). For 60 dB, the signal is 1,024 times
larger than the noise.

4.2.3 Minimum Illumination

Noise sensitivity is impossible to eliminate completely, so as the signal
level drops (i.e., as the light level of a scene drops) and the noise level
remains the same, the signal to noise ratio drops, and noise will begin to
appear in the picture. As the light level continues to drop, the scene looks
progressively worse. Although there is no point at which a picture
suddenly becomes completely unusable, there does come a point where at
least half of the people trying to view it would consider it too annoying to
watch or would be unable to discern facial or even other large features
easily. Somewhere before this point, equipment manufacturers are said to
measure the light level of the scene and claim that the video equipment has
that certain minimum illumination requirement, a lux level. In general, the
color carriers and color receptors require more light to function properly,
thus yielding significantly higher lux ratings than the average black and
white camera. For example, a color camcorder's sales ad may refer to a "3-
lux" camcorder, while a black and white camera's sales ad may claim it is
a 0.5 lux camera.

Unfortunately, these statements of minimum illumination leave plenty of
room for confusion. First, the lux is not a common term among
Americans; what does lux mean? Once lux is defined, it is not
straightforward to judge lux levels. Although a user can recognize that one
camcorder presumably needs less light than another to function properly,
he cannot tell what the relative difference is between light levels (e.g.,
between 3 lux and 20 lux). Finally, the user does not know what quality
can be expected at a specified lux level.

Since lux is at the center of this discussion, it is important to get some
feeling for what a lux is. Lux is a measure of illumination that is used
within the International System of Units (i.e., the Metric System). A lux is
defined as one lumen per square meter, where the "lumen" is a well-
defined measure of light power. In other words, when a lumen worth of
light is uniformly distributed across an area of one square meter, the light
level of that area is one lux. The foot-candle is analogous to the lux but
uses dimensions that are more familiar to Americans. The foot-candle is
equal to one lumen per square foot. Since the light source is the same for
both measures, the only difference between the values of lux and foot-
candles has to do with the areas of the illuminations (square meters and
square feet). With one meter equal to approximately 3.281 ft, one square
meter equals 10.76 ft2. This means the illumination of a foot-candle is
more than 10 times brighter than the lux since the same amount of light is
concentrated over a much smaller area. One foot-candle equals 10.76 lux.

But how does a user determine what his typical lux levels will be? Like
resolution, the answer comes after thinking about the conditions in which
video equipment is likely to be used. Table 9 gives some average reference
numbers for various outdoor and indoor conditions. These "rough"
numbers can be used as rules of thumb from which to gauge typical
illumination requirements. In general, it appears that outdoor daytime
applications and indoor applications that have a normal amount of
artificial lighting can be accommodated by most off-the-shelf cameras/
camcorders. Surveillance situations that occur at night or in very dimly lit
locations indoors will require special, low-light cameras. True low-light
camcorders are not available; however, some can be supplemented with
light amplifiers to achieve low-light capability. One example of this is the
Astroscope series of products from Electrophysics Corporation. In
addition, some camcorder manufacturers (e.g., Sony, Panasonic) provide a
feature in some of their models that disable the color channel in low-light
situations, therefore reducing noise and extending the light range of their
camera. While this does not approach the capabilities of light-amplified
video equipment, it is much less expensive, and may improve image
clarity enough to yield success from a bad surveillance situation.

Unfortunately, while manufacturers are willing to tout their light rating,      they are reluctant to provide information on the quality of the video
images at those light levels. Minimum levels of illumination are rarely
given in the context of a resolution figure or SNR. Each manufacturer uses
its own subjective method for determining the least amount of light
required for producing an acceptable image. What then can the prospective
purchaser use to predict performance at minimum light specifications?

Presently, the measurement of signal-to-noise ratio or resolution at the
lighting threshold is really the only way that a user can judge the potential
quality of a video picture at that level. Another valid method, recently
standardized, is based on the way humans see. This method is to record
video from a device under test, digitize the images, and use computers to
extract the same type of information the human eye and brain do, such as
edge information, noise content, frequency content (another measure of
resolution), and many other parameters. These parameters, when used in
conjunction with data accumulated from many similar tests on many
human subjects, allow the computer to judge the quality of an image
produced by a video device in the same way a human viewer would. These
techniques were developed in the laboratories of the Institute for
Telecommunication Sciences and are published in American National
Standard T1.801.03-1996, "Digital Transport of One-Way Video Signals -
Parameters for Objective Performance Assessment."

Furthermore, standardized measures for specifying minimum illumination
will soon be available. These will help users ascertain quickly whether
there is enough light present in some cases to even bother videotaping.
Since many settings requiring surveillance by law enforcement and
corrections officers will be dimly lit, these new measures will be
particularly helpful.

4.2.4 Shooting Below the Light Threshold

At some point, a situation might arise for which the proper equipment is
not available. Most likely, this will be a situation where surveillance must
be conducted in a lower light environment than was originally anticipated.
It may be that the available equipment does not provide a satisfactory
image under the required lighting situation. If such a situation arises,
continue to tape because useful information may be extracted using digital
imaging techniques. An example of these techniques is shown in Figure
12.

Figure 12 contains two images. The image on the left was obtained
directly from videotape. The tape was recorded in a low-light situation in
the laboratory, with the camera pointing at a head-and-shoulders type
picture. The ambient light level was 0.5 lux, and the manufacturer rated
the camera at 4.0 lux. Thus the experiment was carried out at one-eighth of
the minimum light level for the camera. It is obvious that no useful
identification can be made directly from the image. The image on the
right, however, is useful for identification. How do you get from one to the
other?

The image on the right is composed of individual images that have been
averaged together to reduce noise and increase the signal level. Thirty
consecutive frames (1 second worth) of video were used to compute the
averaged image. Since the noise is random, it tended to cancel itself out
over the 30 frames, leaving a reasonable image. There are two conditions
that must be met for this technique to work. The first has already been
mentioned: the noise must be random, as is typically the case when light
levels drop. Second, there must some signal there to recover (i.e., a
recording in total darkness just will not work). Finally, there must be some
mechanism for compensating for any motion that the subject had within
the frame during the interval being averaged. This ensures that the image
of the subject lines up perfectly when the frames are averaged, making for
a clearer picture.

The averaging can be accomplished in several ways. First, it is important
to get the images into a computer to be processed. In this case a Sony
digital camcorder (model DCR VX-1000) was used with a Canon Video
DK-1 Video Capture Kit. Once imported, the averaging can be done with a
number of software packages, including high-powered computational
packages like Matlab [registered trademark] and IDL [registered
trademark] , or graphics art packages like Adobe [registered trademark]
Photoshop [registered trademark]. More detail on the process is given in
Appendix B.

4.2.5 Color Accuracy        Order your surveillance system

Color is another way to judge quality. Even when illumination is sufficient
to allow a camera to operate, there may still be built-in errors in the
camera's color generation circuits that sometimes make its colors appear
less than true. Specifically, problems arise with the phase and amplitude of
the color part of the video signal. The phase of the color signal represents
the hue of a color, and the amplitude carries the saturation information.
Hue is defined as the particular shade or tint of a given color, but also has
color as a synonym.

Examples of hues that may be used in everyday conversation are red,
greenish, and blue-green. Saturation refers to the amount of pure white
mixed in with a given hue. A hue that has no white mixed in is said to be
100 percent saturated, while a hue that is half white is 50 percent
saturated. For example, red and pink are the same hue, but red is 100
percent saturated and pink is more like 50 percent to 75 percent saturated;
pink is just red with some white mixed in. Along with brightness, hue and
saturation completely define all colors available within the constraints of
the video system. To test the color accuracy of a piece of video equipment,
its output is plugged into a device (vectorscope) that can isolate the color
information (hue and saturation) from the brightness (luminance)
information. The hue and saturation are then displayed and inspected
separately. Color accuracy readings are given in degrees and reflect the
deviation of the observed color from the "standard" color. The lower the
deviation figure, the better. Cameras with excellent color accuracy may
have accuracy numbers for colors within 5 degrees.

It would be an impossible and unnecessary task to check all possible
colors and combinations of colors; fortunately, it is sufficient just to check
red, green, blue, cyan (blue+green), magenta (red+blue), and yellow
(green+red). Because of this, a standard chart is available that contains just
these colors - called the "color bars" chart. Sometimes color bars are
broadcast by a TV station after they go "off the air" for the night. If a
camera is aimed at a color bar chart (under appropriate illumination), then
its output signal will be ideal for testing with a vectorscope, and it can be
determined whether the camera's reproduction of blue, for example, is
accurate enough for its intended application.

The reason for investigating the color accuracy of a camera or camcorder
is to ensure that colors accurately divulge the race, hair and eye color,
clothing, and other distinguishing characteristics of a suspect or other
interviewee. It is also essential to maintain color information about pieces
of evidence such as paint chips or blood and to identify quickly and
accurately health problem symptoms, such as cyanosis, in a victim.

4.2.6 Maximum Lens Aperture

This characteristic of cameras and camcorders tells something about how
good the lens is. The better a lens is, the faster it gathers light for
conversion into electrical signals and the easier it is for the rest of the
camera or camcorder to maintain a noise-free image, since there is more
light coming from the scene than there would have been with a slower
lens. If most of the perceived applications involve nighttime or twilight
conditions, it is important to obtain the fastest lens available (within the
established budget).

Most cameras/camcorders have the maximum aperture marked on (or
near) the lens as an "f-number." It may be shown, for example, as "f/1.4,"
"1:1.4," or simply "1.4." Speed in a lens, as determined by the aperture,
provides an indicator of how long it takes to expose the video pick up (or
any type of film). With an f/1.4 lens, the light can be one-eighth as bright
as with an f/4 and still allow the same length exposure. For the same
brightness of light on the subject, the shutter speed can be shortened by a
factor of eight. For the police photographer, lens speed is of the greatest
value in a covert surveillance situation where there is little natural light
and supplementary illumination cannot be used. A lower f-number may be
the difference between getting a picture and not getting it.

Table 10 lists standard f-numbers and compares the relative brightness
requirements of each. The table uses f/1.4 as a typical "best" limit for
lenses, however, some cameras offer an f/1.2 lens. An f-number of 1.0 is
the theoretical lower limit for standard lenses.

For most surveillance cameras, it is fairly easy to cover applications
involving different lighting conditions by buying more than one lens.
Lenses can be switched on and off the camera body, even in the field.
Camcorders are not so flexible. The lens that comes with the camcorder
cannot normally be removed.[7] If two camcorders are essentially the
same except for the maximum aperture, it may be prudent to choose the
one with the lower f-number, especially if both show the same lux
specification.

4.2.7 Minimum Focusing Distance

Almost all of today's common camera and camcorder lenses focus as close
as 2 1/2 feet to 3 feet. This is suitable for satisfying most needs of law
enforcement and corrections, except for acquiring some forensic footage.
If there is a need to record extremely close footage of an object that will
serve as evidence, the 'macro' function of the lens should be engaged, if
available. This effectively switches the lens into another mode that has a
focusing range from about 2 1/2 feet right down to zero - which is right at
the lens. In this mode, focusing is typically accomplished with the zoom
controls and with more difficulty.

4.2.8 Zoom

Although such a function is considered a "special" lens on a photographic
camera, it is the norm for video cameras and camcorders. Usually a lens
will come with the unit and be designated, for example, an "8:1" zoom.
This means that if the lens is zoomed to its wide-angle limit (that which
makes the subject look the farthest away), then zoomed to the other end
(as close as possible), the subject will appear eight times as close. This
extreme is called "telephoto." While not true for all lenses, in most
camcorders, zooming to the telephoto end of the lens also reduces the light
transmission (i.e., f-rating) through the lens. In spite of this restriction,
having a zoom feature means the camera or camcorder can be used in a
wider range of situations than would otherwise be the case.

When zoom functionality is solely dependent on the lens, it is considered
an optical zoom. Some of the new digital camcorders have a digital zoom
feature in addition to the optical zoom, in which they use a subset of the
elements of the CCD video pick-up device and enlarge that subset to cover
the full screen. When reporting zoom ranges, manufacturers typically
multiply the optical zoom and the digital zoom features. For example, a
camcorder with a 12x optical zoom lens and a 5x digital zoom feature
would be touted as a 60x zoom. That is, the image of an object fully
zoomed in (magnified) will appear 60 times larger than if the camera were
fully zoomed out. Remember, however, that much of the increase was due
to sampling a subset of the CCD, thus reducing the overall resolution of
the image.

4.2.9 Autofocus

A number of consumer-grade cameras and camcorders on the market offer
the user the choice of either manually focusing on a subject or allowing
the video device to automatically focus. In some cases, the user has no
choice - the equipment comes with only a manual or auto-focus lens.
Whether the auto-focus is optional or mandatory, it is important to realize
what its capabilities (and limitations) are before making a selection.

Two methods are typically used in auto-focus cameras and camcorders -
contrast-maximization and infra-red ranging. These methods are described
below.

Contrast Maximization (CM)

Imagine, as in the discussion on resolution, a series of vertical lines
alternating between pure black and pure white. If this scene is viewed with
a lens system that is in focus, the boundaries between the two types of
regions will be distinct. If the eye scans from left to right, it sees light
levels alternating in sequence between a very low light level (a black
region) and a very high light level (a white region). If the lens is
completely out of focus, all that will be seen is a large area of uniform
medium gray (a medium light level). In fact, for any given scene, the lens
setting that is in focus also produces the maximum contrast in light levels.
Cameras scan pictures as described above but see pictures in terms of
electrical levels. A camera can find the best focus by changing the lens
until the difference between the highest voltage and the lowest voltage in a
scene is maximized. This principle has been exploited to allow camcorders
to find the "best" focus automatically and is called "contrast
maximization."

Infrared Ranging (IR)

In this radar-like system, an infrared emitter, typically located next to the
camcorder's lens, transmits a pulse of light that is not in the range visible
to humans. An infrared light sensor then waits for a reflection of the
original pulse and notes how long it took for the reflection to return.
Knowing how fast the pulse travels, the system then calculates the
distance to the object that reflected the light pulse and adjusts the lens
apparatus accordingly. These calculations and adjustments are done from 5
to 10 times per second.

In implementing both the CM and IR methods, engineers had to answer
many questions. For example, in the case of the CM camcorder, the true
focus could be anywhere in the range of the lens (or too close). While
sweeping from nearest to farthest, the contrast might not simply increase
until focus is attained, but instead, it may increase a little, then decrease,
and then increase a lot. How should the camcorder decide whether to find
an even greater contrast, or to stand pat on its current decision? If it stays
where it is, it might not be in focus, and then just stay there, out of focus,
forever. If it is in focus but goes hunting for a better focus, it might simply
waste time, losing valuable information by recording out of focus until it
comes back or stays elsewhere incorrectly.

In the case of the IR focusing camcorder, the pulse should spread as it
travels outward, away from the camera and toward the subject to be
videotaped. If it does not spread but stays thin as a pencil, it could focus,
for example, through to the other side of library bookshelves (when there
is an opening through and things and people on the other side can be seen)
when the books are the intended subject. If the beam is too wide, the
camera could focus on the books, when it is the people on the other side
who should be monitored. In another scenario, what if the user of the
camcorder is by a chain-link fence, and something else is on the other
side? Should the point of focus be the chain-link fence or the object on the
other side? Both cases are possible, but how does the user inform the
camera what's on his/her mind? He/she does not with auto focus!
Manufacturers made decisions for the user.

Regardless of how the manufacturers ultimately decided, it is imperative
that the user understand what they decided to do and how that impacts the
effectiveness of auto-focus for certain video applications. Some of the
advantages and disadvantages of the two technologies are given below:

IR focuses on glass, whether you want to or not. This makes observation
of people in houses and vehicles very difficult.

CM looks for the greatest difference between black and white. If the room
is relatively dark, it takes longer to decide that a given lens setting is right
or wrong. It could take up to 30 s to focus in a well-lit indoor room when
the subject is of low contrast already.

IR systems can focus perfectly on a subject even if there is absolutely no
visible light. This does not mean the camera can operate and/or record
images in pitch black surroundings. This means the camera can be focused
and ready to go, and once the light level reaches the minimum illumination
the camera requires, it will operate. An IR autofocus camera, then, could
be placed in a dark room and record happenings when a lamp was turned
on or some sunlight shone in. From bright light to little light, the IR
focusing ability is not progressively diminished as the light grows dimmer.

CM uses only the center third of the frame when calculating focus;
therefore the zoom interacts with the auto-focus. CM maintains focus
when moving from wide-angle to telephoto but loses it badly in the other
direction.

Regardless of whether the auto-focus mechanism is CM or IR, depending
on the particular camera, the auto-focus can be swift or slow. The swift
ones tend to hunt indecisively once they get close to focus, whereas the
slow ones take longer to get there but are very accurate and stable once
they decide they are focused.

Linked in with all of this is the auto-iris, the part of the camera that adjusts
for changing light levels. As the iris opens and closes, the depth of field
will change, making parts of the scene that are not at the same distance as
the subject go in and out of focus. While the CM types will be affected by
this, the IR types do not appear to be.

4.2.10 Shutter Speed Control

The fact that most video devices in the United States produce frames at the
rate of 30 per second means that if a subject moves considerably in 1/30
second, it will appear blurred under normal circumstances. One of the
features available (with the proper device) that can solve this problem is
shutter speed control. Just as with a photographic camera, higher shutter
speeds reduce the amount of time the shutter is open and decrease the blur
seen during that time. The payback is the faster the shutter speed, the more
light that must be available; however, with a subject illuminated by full
sunlight, shutter speeds of one 0.0001 second are possible with
uncompromised quality.

--------------------------------

5. The Ergonomic Aspects of Equipment

One of the experiences people have had with new video equipment is they
could not get it to work. No, this is not another story relating to poor
workmanship, missing parts, or bad information from the salesman! In the
vast majority of cases, there was nothing wrong (theoretically) with the
equipment. Problems arose because users did not know what to do; the
equipment was not straightforward to operate; technical manuals were
incomplete, misleading, or confusing; or the equipment controls could not
physically be moved or positioned. These were all ergonomic problems.
(Ergonomics is the science concerned with the characteristics of people
that need to be considered in designing and arranging things. That is, how
should something be made so that people will interact with it effectively?)
A video product can have the best specifications and features in the world,
but if no one can easily use it, it has no real practical value.

This section addresses a few of the "nitty-gritty" items that may be
forgotten during the selection process for video surveillance equipment. If
specifications, features, and cost all fall out as about equal, one of these
items may be the deciding factor. Even if the video units mentioned below
did not have any drawbacks in specific categories, it still would be prudent
to look at those kinds of categories when actually contemplating a
purchase.

5.1 Time Needed to Learn Basic and Advanced Operations

With a technical manual as a guide, it only took about 5 or 10 min for
ordinary people (non-experts) to get the various video units described in
this guide working. (That is, if the batteries came already fully charged
with the camcorders). The Sony DXC-M7 camera took somewhat longer
because it is a more complex piece of equipment (e.g., more function
switches, feature controls, and connectors). An officer who had never used
a camera or camcorder before would be ineffective if forced to use one
"cold" in a pressure situation, but 2 h of use over a couple of days
probably would allow that officer to use all of the basic features
adequately and to record valuable information. To learn and use advanced
features, such as toggling sensitivity gain and autofocus, adjusting white
balance, connecting a 10 W lamp to the camcorder, and using the macro
feature of the lens, some weeks of use would be required. In addition,
there are some things an artist can do with skillful control of the
camcorder that your average operator will never be able to do. This
personal element is no more of a problem with certain officers than trying
to photograph evidence with a standard 35 mm single lens reflex camera,
however.

5.2 Controls - What Kinds are Better?

This subject is general to cameras, camcorders, televisions, monitors, and
VCRs. Even in this realm, things have changed a lot since the days when
TVs were powered on with a loud mechanical click. This is mostly due to
solid state electronics, but also to materials science. Since silicon switches
use so little energy, it is not uncommon now for a part of an electronic
instrument to remain on even when it has been turned off. Turning it back
on consists of moving a switch that closes an electrical contact that tells
the part of the instrument that remains on to turn on the rest of the
instrument. Since the switch is just a contact, it can be made quite small.
Unfortunately, some manufacturers have gotten overzealous in their desire
to show off just how small they can make their switches, and the result is
full-size camcorders that have switches that are too small to use
comfortably with just fingers, not to mention gloves.

The most common types of controls are buttons, sliders, knobs, and
switches. (See figs. 3, 9, 10, and 11.) Switches are generally of the variety
that can be in one of two positions and stay there. More popular than
switches are buttons, knobs, and sliders. Buttons most often are just
electrical contacts that electronically toggle functions: the first time you
press it the auto-focus turns off; the next time it turns back on. Of course,
if these toggle buttons are too large, it becomes possible to bump them
when it is not desirable to do so. It could be disastrous if such a button
controlled the power for the entire camcorder or for the cassette eject
mechanism. These functions are usually protected by a slider, the function
of which is invoked by moving a spring-loaded control to the side
momentarily for electrical contact. Sliders return themselves to a home
position upon release. Since pressure straight upon their surfaces does not
bring them to operation, they are safe for the more important functions.
Knobs are most commonly used to allow the user to select one of many
options. One example of this is the power-on knob of the Sony DCR VX-
1000. This knob allows selection of VCR or camcorder mode, as well as
various levels of automation.

An example of a more complicated control that can be found on some
devices is the power control on the Sony CCD-V99 camcorder. It has a
center position and slides to one side to use the device as a camcorder or to
the other side to use it as a VCR. Before moving it from its center
position, however, a tiny green button in the middle of the sliding switch
must be depressed to unlock it. It has proven to be a formidable task even
with bare fingers.

Even simple controls like buttons can be made difficult to operate by
placement. Many times frequently used controls are placed behind covers
or doors. This slows access to those controls. One extreme example of
poor placement are the menu controls on the Sony DCR VX-1000. These
controls are behind the battery compartment door, on the side closest to
the hinge. This makes the buttons awkward to reach without the added
complication of having to hold the camcorder at eye level to see the menu
in the viewfinder and having the battery door continually bumping into
your chin.

Monitors and televisions are used almost exclusively indoors or at least
not in sub-zero outdoor weather. Since they are relatively large, there does
not seem to be a leaning by designers to give them unnaturally small
switches. Operation is rarely an ergonomics problem.

In general, understand that modern equipment will have many controls to
maximize functionality, and they may need to be somewhat small to fit
them all logically onto a control panel. Beware of controls that are smaller
than they need to be - especially if you may need to use the equipment
with gloves or in tight places.

5.3 Camcorder Use with Gloves and Other Heavy Clothing

Most of the hand straps that come manufactured with camcorders can be
adjusted to fit large hands, even ones sporting gloves that are not too thick
(e.g., driving gloves or work gloves). Shoulder supported camcorders
(most standard size VHS units) and cameras may not have a hand strap but
an area built through the device that is intended for use as a grip. The size
of this grip may not be adjustable, and this should be taken into account
when considering purchase. The low end of the operating temperature
range is usually specified to be above the freezing point of water (32
[degrees] F)anyway, and extremely thick and heavy gloves probably
would not be needed at this temperature. The other consideration when
using thick, heavy gloves is the size of the buttons and knobs on the unit.
This should not be a problem either if gloves are not too thick, given the
typical button and knob size.

5.4 Weight and Handling Versus Steadiness When Operating

Weight is mostly irrelevant when dealing with devices intended to sit on a
shelf, such as most VCRs and monitors. Some monitors can be transported
between sites, but system design options should not include having a
monitor strapped to a human assistant. For cameras, portability, weight,
and handling are a significant issue, because manufacturers always try to
compromise between quality, manufacturing cost, and consumers' desires
for something small and easy to operate. This is even more true when you
try to design a tape recorder into the same small enclosure as the camera.
Such a device is called a camcorder. As far as weight goes, it is desirable
to keep it low, but stability will suffer if weight is insufficient to keep the
camcorder steady when the operator moves. Muscles sometimes shake a
little when asked to remain perfectly still. At the other end of the extreme,
even aside from the obvious discomfort of carrying around a camcorder
weighing 15 lb, excessive weight can make muscles shake just from the
sheer effort of supporting it after awhile. Somewhere in between is the
ideal weight for a particular operator.

In addition, since 8 mm videotape is so much smaller than VHS videotape,
most camcorders employing it are smaller and lighter than their VHS
counterparts. They are carried in front of the operator's body and face,
whereas VHS models are typically carried on the shoulder, as simple video
cameras are. The compact version of VHS, called VHS-C, can be carried
in front of the operator like 8 mm camcorders also, since their videotapes
are much smaller than standard VHS videotapes. If the camcorder is
carried in front of the operator, the device is typically lighter than one
carried on the shoulder, but it has only the operator's two hands to support
and steady it. If the camcorder is carried on the shoulder, the device is
typically heavier than the 8 mm or VHS-C varieties, but the shoulder
support is both quite strong and very stable or steady.

All cameras and camcorders come with a screw mount on the bottom for
attaching to a tripod. A tripod can serve to simplify surveillance, as the
burden of supporting the machine is moved to the tripod, and there is no
risk of a human operator wavering off target.

5.5 Equipment Compatibility

If your entire contingent of video equipment is of VHS format and you
acquire one 8 mm camcorder, then when you want to view the tapes you
have recorded with the 8 mm camcorder, you will have to use the
camcorder for playback. In addition, within the VHS universe, if you
acquire a VHS-C camcorder, you will need an adapter to play its tapes on
standard VHS systems. These adapters are usually included with VHS-C
systems.

5.6 Helpful and Useless Features

The ability to switch off automatic features is invaluable if an operator is
skilled in the use of a particular piece of video equipment. There is always
a situation where autofocus is undesirable or ill-suited or the operator must
force the iris open to gain detail for the features of a face against a brighter
background.

Since camcorders are already quite a mature product, most of the features
that are required to obtain quality images are available on almost every
model. Manufacturers try to distinguish their products through the addition
of features that can be generally considered useless for video surveillance
applications. Part of the reason for this is in most cases it costs the
manufacturer so little to include features (e.g., titling, strobe effects, artful
fades or dissolves from one take to another) that they are just installed as a
matter of course. This is especially true as microprocessors evolve and
drop in price. It is conceivable that, since the complexity of the device is
increased to incorporate these features, the chance is increased that an
officer not so experienced with video equipment might press the wrong
button and actually lose the ability to accurately record information.

5.7 Viewfinders

Are some viewfinders bigger and better than others? Until recently,
viewfinders were almost always just a small (about 1 in diagonal)
monochrome CRT connected electrically to the body of the
camera/camcorder. More recently, color LCD viewfinders are coming to
dominate the camera and camcorder market. These viewfinders are
typically mounted within an enclosure that can swivel up and away from
the body to allow the user to get the camera lower for shots of children or
to shoot under a fence, for example. The viewfinder is made comfortable
to place against the user's face by including an eyecup of very flexible
rubber molded to approximate the average user's facial contours. Since the
user is actually placing his eye up to the eyecup within 1 in or 2 in of the
CRT within the viewfinder and the eye can't focus at that distance, a lens
is provided within the viewfinder between the eye and the CRT. This lens
can be adjusted to match the natural focal length of the user's eye so that
extended use of the viewfinder is comfortable.

In addition to viewfinders, many consumer camcorders are available with
a 2-in to 4-in LCD monitor that flips out from the camera body to tilt and
swivel. Many camcorders with an LCD monitor also have a viewfinder,
although some models have totally replaced the viewfinder with the
monitor.

It is not a straightforward choice to select between the types of
viewfinders. The small viewfinder must be held to the eye but the required
stance is stable, and the aiming motion is quite natural, ensuring the
intended subject gets recorded on the tape. The LCD monitor allows more
flexibility in holding the camera and a larger display on which to read all
the information provided by the camcorder. It also allows the videographer
to interact more directly with those around, making the subjects more
comfortable with the presence of the camera. Sacrificed are a little bit of
stability and precision. If possible, a camcorder with both would be
desirable and provide the most flexibility.

5.8 Battery Life and Replacement

Rechargeable batteries[8] supplied by the camcorder manufacturer are
intended to last for 2 h - the length of time available for recording on one
videotape (8 mm and VHS). When one tape is completely full of recorded
material, tapes and batteries can be swapped simultaneously, and then the
expended battery can be connected to the recharging unit so it can be ready
in 2 h. Multiple batteries are rarely supplied with the camcorder, so it will
be necessary to specify extra batteries at the time of purchase of the
camcorder. Even though extra batteries are not included, some units
provide charging capacity for two other batteries while another is being
used.

Some of the more compact camcorders carry their rechargeable batteries in
a compartment under the hand strap. Thus, during camcorder operation,
the user's hand wraps around the battery compartment. This can be a good
attribute, because the warmth that the hand provides also keeps the battery
warm and, electrically, more potent. This can also be bad, because it can
be less than convenient to try to exchange batteries in a hurry when a panel
has to be removed and batteries have to be removed out from under the
hand strap.

5.9 Tapes - Cost versus Quality; Problems Reading Tapes

It is commonly felt (and consumer product testing firms have found) that
tape is tape is tape, and all tapes record the full bandwidth of their
respective formats (e.g., VHS, 8 mm, S-VHS). There does not seem to be
a difference even between regular grade and "high-grade" tapes. Also,
there is no good reason to pay extra for tapes designated as "hi-fi," since
any tape can record high-quality sound in a VCR that records in the VHS
hi-fi format.

The defect that is found on videotapes manifests itself as "dropout," where
the signal is lost temporarily, so the playing machinery must
resynchronize. It is the frequency of these dropouts that determines the
relative quality of videotapes. It is worth noting that the average viewer
does not notice most dropouts, although this is little consolation to work as
critical as law enforcement and corrections. As a general rule, it may be a
better approach to buy brand-name tapes on sale than to buy off-brand
tapes that may not have satisfied the same types of quality manufacturing
standards.

As far as reading tapes, there should not be any problems except for those
associated with the environment. The heads that read the tape are actually
dipoles mounted in the surface of the rotor, and the helical rotor actually
does not touch the tape being transported across it at an angle but forces a
film of air between its own surface and the tape because of friction and
high rotation speeds. Moisture particles in the atmosphere (from simple
humidity or outright rain) can be larger than the gap between the rotor and
the tape, causing drag and improper operation. This can be sensed and
relayed to the operator, usually with a "DEW" indicator, such as an LED.
When this indicator appears, remove the battery and let the camera sit
(with all doors open) in a dry spot for a couple of hours (or overnight)
before trying to use it again. This should give the moisture enough time to
evaporate.

5.10 Maintenance for a Machine with Tape Heads?

While some newer camcorders and VCRs have self cleaning heads, head
cleaning is one of the most common maintenance tasks for these devices.
Here is a paragraph from one manufacturer's operating instructions:

Cleaning the Heads: It is recommended that head cleaning be performed
by a qualified service technician. Please contact your nearest Service
Center. An alternate solution is to obtain a head-cleaning cassette. There
are many types of cleaning cassettes, so be sure to follow the cleaning
instructions carefully. Excessive use of the cleaning cassette could shorten
head life. Use this cassette only when a head clogging symptom occurs.

Cleaning heads on any helical scan device, whether VCR or camcorder, is
almost a judgement call. They do not need to be cleaned exceedingly often
unless the work they record or reproduce is critical. When cleaning is
necessary, it can best be done by disassembling or reaching in with special
equipment - in other words: professionally. It can be done with head
cleaning tapes, which consist of an abrasive material manufactured into a
cassette just like a standard videotape. They are first wet with a head
cleaning fluid and then "played" in the camcorder or VCR. Sometimes,
because of their abrasiveness, they are not recommended by the
manufacturer of the camcorder or VCR, and sometimes they just do not do
the job very well anyway.

5.11 Documentation/Instructions

Camcorders are manufactured exclusively in foreign lands. Unfortunately,
manufacturers believe, for some reason, it is not necessary to hire native
speakers from target market countries to write or assist in the writing of
documentation for these pieces of equipment. The result can sometimes be
confusing and frustrating.

Documentation has been found to be complete. The technical writers
working for the manufacturer try to make documentation complete for a
unit by binding together manuals for several closely related units. This is
not enough, however, since the writing is often poor in grammar and
clarity. If a camcorder does NOT have a particular feature, such as the
capability of turning off autofocus, it probably will not say so in the
manual. The obvious intent is not to highlight a lack of something in the
product, but it might be beneficial to the user to know it as soon as
possible.

--------------------------------

6. Summary

With all the advances in videography today, there will come a day in the
not-too-distant future when still photography will no longer be the
preferred technique for recording data for most law enforcement and
corrections needs. As the resolution and electronic shutter speeds of video
equipment continue to improve and the costs of video units are reduced
even further, the current advantages of conventional photography will
diminish. Also, digital video and multimedia computing could have a
significant impact on the future of video surveillance and how the data are
gathered and processed. That is why the basic concepts of this guide are
important to both imminent purchasing decisions and planning in
anticipation of new technologies.

This guide has attempted to convey the many aspects of video in enough
detail to allow a fundamental understanding of technical parameters and
how they relate to law enforcement needs. It is hoped, however, that the
discussions of the guide will have stimulated readers to conduct
subsequent investigations into the ever-changing capabilities and
applications of video gear. Only by having a clear recognition of what the
potential benefits are, can those in the law enforcement and corrections
communities hope to take advantage of the ongoing video revolution.

--------------------------------

7. Glossary

amplified light - An attribute of a camera or other video device indicating
use of a special module to amplify ambient light before it gets to the pick
up unit.

amplitude - The voltage level of a signal. Could be relative (e.g., peak-to-
peak for ac signals) or absolute (for dc signals).

aspect ratio - In facsimile or television, the ratio of the width to the height
of a picture, document, or scanning field. NTSC television has
standardized the aspect ratio at 4:3 (i.e., the picture is wider than it is high
by a factor of 1 1/3). If an image is not reproduced at the intended aspect
ratio, objects in the image are distorted.

automatic iris control - An automatic control that regulates the amount of
light that reaches the video pickup unit.

auxiliary jacks - Any of a number of connectors that a piece of video
equipment can have to allow it to be connected to and interwork with other
equipment.

bandwidth - The difference between the limiting frequencies within which
performance of a device, in respect to some characteristic, falls within
specified limits. An analogy to bandwidth might be the width of a street or
a highway, where each lane is a radio frequency.

battery - A device used for storing energy until it is required for use by a
piece of equipment. Enables equipment to work without being plugged
into a wall outlet.

battery memory - In rechargeable batteries, refers to the tendency of some
batteries to "remember" the level to which they were charged or
discharged, reducing the overall useful storage capacity of the battery.
(See NIJ Guide 200-98, "New Technology Batteries Guide," for more
information.)

black balance - See white balance.

blue-only control - A switch that turns off the red and green electron guns
in a monitor. This allows for the monitor to be calibrated based on the
signal from the blue gun only.

book mark - A feature of camcorders and recorders that allows the user to
quickly find the end of previously recorded material so that additional
recording can resume from that point.

brightness - A qualitative attribute of visual perception in which a source
appears to emit a given amount of light. In monitors, overall brightness is
dependent on the high-voltage level and the dc-grid bias.

broadcast quality - A generic descriptor indicating a piece of equipment is
of sufficient quality to be used regularly by the broadcast television
industry. Typically, the requirement is that resolution be greater than 450
TVL.

CCD (charge coupled device) - These tiny light-to-electric-charge
transducers are placed in rectangular arrays on silicon wafers and used as
video pickup devices instead of electron tubes. The signal is read out from
the array sequentially from side-to-side and top-to-bottom to determine
one video frame.

chrominance - In color television, that signal or portion of the composite
signal that bears the color information.

clarity - A qualitative term generally referring to the combination of
resolution, contrast, and color accuracy.

CM (contrast maximization) - a technique for autofocusing cameras and
camcorders based on maximizing the contrast of the video signal.

color - Having a non-white spectral characteristic.

comb filter - a filter which helps to minimize the loss of resolution and
reduce streaking and wavy edges on fine patterns. Common in middle
range to high-end television displays and monitors.

contrast - In display systems, the relation between (a) the intensity of
color, brightness, or shading of an area occupied by display elements, a
display group, or a display image on the display surface of a display
device and (b) the intensity of an area not occupied by a display element, a
display group, or a display image. For a monitor, contrast is determined by
the peak-to-peak amplitude of the video signal.

counter - In cameras and recorders, counters are used to keep track of tape
position between start and finish. Counters can be in arbitrary units, time
counting up, or time counting down.

CRT (cathode ray tube) - the vacuum (electron) tube that generates an
image in a television monitor using cathode-ray electrons.

dB (deciBels) - 1) one tenth of the common logarithm of the ratio of
relative powers (P), equal to 0.1 bel. The formula is given by dB = 10
log[10] (P[1]/P[2]). 2) One-twentieth of the common logarithm of the ratio
of relative voltages (V) or currents (I), equal to 0.1 bel. The formula is
given by dB = 20 log[10] (V[1]/V[2]) for voltage and dB = 20 log[10]
(I[1]/I[2]) for current.

dichroic lens - A lens in a camera which splits the incoming light into the
three primary colors (red, green, and blue) so they can be picked up by
separate CCDs or different areas on one CCD.

digital zoom - A relatively new feature of digital cameras whereby they
use only a portion of the pickup device and magnify the image to fill the
full frame.

distance - The position of the subject relative to the camera.

DSO (digital storage oscilloscope) - an electronic test instrument used
primarily for making visible the instantaneous value of one or more
rapidly varying electrical quantities as a function of time or of another
electrical or mechanical quantity. Its storage function allows several values
to be recorded (and displayed together).

DV in/out - IEEE 1394 (also known as "FireWire") interface available on
digital camcorders.

dynamic contrast control - An automatic control to maximize the contrast
of a scene. Generally, use of dynamic contrast control produces an
improvement in overall picture quality.

electron beam spot size - The diameter of the focused electron beam that
causes the phosphor on a monitor screen to fluoresce.

edit controller - A jack on a piece of equipment that allows it to be
precisely controlled by another device for the purpose of editing tapes.

electron tube - A vacuum tube designed to focus and direct beams of
electrons. A common type of electron tube is a television picture tube (i.e.,
a CRT).

electronic shutter - Use of electronics to simulate placing a shutter in front
of a video pick-up device.

environmentally robust - A manufacturer's subjective claim that their
equipment can operate in a variety of temperature, humidity, lighting and
physically abusive conditions.

fade - A non-abrupt interruption of the signal. In video, generally refers to
a graceful transition from one video signal to another.

filters - In electronics, a device that transmits only part of the incident
energy and may thereby change the spectral distribution of energy.

flying erase head - In camcorders and recorders, a recording technique that
allows for a single frame to be erased from a video tape and then
immediately replaced with a frame from another source. This allows for
smooth transitions between scenes.

focus - The mechanism used to ensure that the scene produces a sharp
image on the video pickup device.

gain-up - A control to increase the gain on the output of the video pickup
device in low-light situations.

headphone jack - On video equipment, this is usually a 1/8 in stereo phono
jack.

high definition television (HDTV) - Television that has approximately
twice the horizontal and twice the vertical emitted resolution specified by
the NTSC standard.

high-speed shutter - A physical or electronic shutter that operates at faster
than 1/60 s.

hue - The visible spectral content of an image or part of an image, which
depends on the phase angle of the chrominance signal. The phase is varied
with respect to a color synchronizing signal by a "tint" or "hue" control.
This control is subjectively set for the correct hue of any known color on
the screen (e.g., green grass or blue sky), then all other hues are
automatically corrected, since the color synchronization holds all hues in
the proper phase with respect to each other.

image stabilization - A camcorder or camera feature to reduce the visible
effects of shake and wobble introduced by hand-holding the camera. Two
techniques are currently used to accomplish this. The first is through the
use of a deformable prism. As the camera/lens detects shake and vibration,
the prism is reshaped to provide stability to the image. The second is to
electronically remove the effects of shake and distortion by modifying the
output signal from the pick-up device.

index - A feature that "marks" the videotape each time recording is started,
enabling the user to easily find a particular recorded section of tape.

infrared light - The region of the electromagnetic spectrum bounded by the
long-wavelength extreme of the visible spectrum (approximately 0.7 mm)
and the shortest microwaves (approximately 0.1 mm).

infrared playback - See wireless playback.

inputs - The types of signals that a device can receive, and the connectors
through which those signals are received.

intensifier - A device placed in front of a camera or camcorder's pickup
device that amplifies available light from a scene.

IR ranging - An autofocus technique that uses an infrared signal to
determine the optimum focusing distance.

iris - The adjustable physical opening that light passes through en route to
the video pickup unit.

intermediate frequency (if) - A frequency to which a carrier frequency is
shifted as an intermediate step in transmission or reception.

LANC - Sony's edit control interface for high-end consumer equipment.

LCD monitor - A viewing device for a camera or camcorder that is based
on liquid crystal display technology and is 2 in to 4 in in size.

lens compatibility - Indicates a camera has many interchangeable lenses,
including interchange-ability with those of other manufacturers.

lens mount - The physical connection between the lens and the camera.
The most common lens mount for video cameras is the "C" mount.

light (1) - In a strict sense, the region of the electromagnetic spectrum that
can be perceived by human vision, i.e., the visible spectrum, which is
approximately the wavelength range of 0.4 mm to 0.7 mm.

light (2) - An attachment for a camera or camcorder to help illuminate
scenes where available light is too low to allow recording of a satisfactory
image.

low light - Low-light cameras typically have published minimum
acceptable light levels between 0.1 lux and 2 lux.

luminance - In color television, that signal or portion of the composite
signal that bears the brightness information.

lumen - A well-defined measure of light power emitted by a source.

lux - The light level incident on a 1 square meter area when a lumen of
light is distributed across it.

macro mode - A special mode for some lenses that allows focusing at
closer distances than normal to provide greater magnification of a small
object or detail on a larger object.

microphone holder - A bracket on a camera or camcorder that allows
attachment of an external microphone.

minimum illumination - The minimum ambient light level (usually given
in lux) required to give the camera a sufficient signal to make an
"acceptable" picture. Each manufacturer has a different definition of
acceptable.

monitor bridging - a mode in which a monitor can receive and display a
video signal and then pass it on to another device without modification.

motion sensor - An automatic sensor in a camera or camcorder that allows
the system to be activated when motion is detected and deactivated at a
specified interval after motion ceases.

multiple heads - In video playback units, multiple heads improve the
image quality during high-speed and slow-motion playback.

multiple mounting holes - For cameras, multiple tripod mounting holes
enable the camera to be balanced atop the tripod to provide more stable
images.

noise - A disturbance that affects a signal and may distort the information
carried by the signal, or, loosely, any disturbance tending to interfere with
the normal operation of a device or system.

noise reduction - Using filtering or digital signal processing techniques to
reduce the amount of noise in an image. Noise reduction figures of 6 dB
are common.

NTSC (National Television System Committee) - denotes the body that set
the original standards for American television and is also used as a
reference to the television standard they published.

NTSC video - The North American standard (525-line interlaced raster-
scanned video) for the generation, transmission, and reception of
television signals. Note: In addition to North America, the NTSC standard
is used in Central America, a number of South American countries, and
some Asian countries, including Japan.

optical zoom - The zoom achieved by a lens.

phase - Of a periodic, varying phenomenon (e.g., an electrical signal or
electromagnetic wave), any distinguishable instantaneous state of the
phenomenon, referred to a fixed reference or another periodic varying
phenomenon. Note: The phase of a periodic phenomenon can also be
expressed or specified by angular measure, with one period usually
encompassing 360 [degrees] (2 p radians).

photo mode - a camcorder/videotape recorder "captures" a single frame of
video and records that one frame for 6 s to 10 s on the videotape,
essentially making a still photo on the videotape.

pixel - In a raster-scanned imaging system, the smallest discrete scanning
line sample that can contain gray scale information. An abbreviation for
picture element.

playthrough - The condition of taking a known input, passing it through a
device, and comparing the output of the device with that known input.

radio frequency (rf) tuner - The part of a circuit that can be adjusted to
resonate at a particular frequency. Allows "channels" to be received from
broadcast or cable systems.

remote control - a device that is detached from the main chassis of a piece
of equipment, yet provides a mechanism for the user to control that piece
of equipment. The two most common types of remote control are wired
and wireless. Wired remotes require a physical connection (via wire) from
the remote control to the main chassis. Wireless remotes typically use an
infrared signal to communicate between the remote control and the main
chassis.

resolution - A measurement of the smallest detail that can be distinguished
by a video system or device under specific conditions.

rf (radio frequency) - any frequency within the electromagnetic spectrum
normally associated with radio wave propagation. Normally, information
signals are modulated to be transmitted at a radio frequency.

RGB (red-green-blue) - pertaining to the use of three separate signals to
carry the red, green, and blue components, respectively, of a color video
image.

RS-170A (EIA-170) - An Electronic Industries Alliance (EIA) standard
describing a black and white television system containing 525 lines in two
interlaced fields at a field rate of 59.94 Hz. This is the basis of the modern,
North American NTSC television system.

saturation - In video systems, the level of color relative to the maximum
handling capacity for that color. The level of saturation is dependent on
the level of the chrominance component of the video signal.

scan rate - The frequency at which the electron beam scans a single line of
an image. This is 15.7 kHz for an NTSC system and can be as high as 100
kHz for computer monitors.

screen size - the diagonal dimension of a display screen (measured in
inches or centimeters). Sometimes part of a display screen may be hidden
behind a plastic housing (i.e., the case of the display), thus causing a
mismatch between the published screen size and the viewable screen size.

self timer - A feature of a camcorder or video recorder that allows it to turn
itself on and/or off at a particular time or time interval.

sensitivity - In an electronic device (e.g., a communications system
receiver such as a television), the minimum input signal required to
produce a specified output signal having a specified signal-to-noise ratio
or other specified criteria.

shutter - A device that opens and closes, allowing or disallowing light to
reach the video pickup device.

signal - Detectable transmitted energy that can be used to carry
information.

SNR - signal-to-noise ratio - the ratio of the amplitude of the desired
signal to the amplitude of noise signals at a given point in time. Note 1:
SNR is expressed as 20 times the logarithm of the amplitude ratio or 10
times the logarithm of the power ratio. Note 2: SNR is usually expressed
in dB and in terms of peak values for impulse noise and root-mean-square
values for random noise. In defining or specifying the SNR, both the
signal and noise should be characterized (e.g., peak-signal-to-peak-noise
ratio), to avoid ambiguity.

speaker - An electrical signal to audio sound pressure transducer.

special effects (special FX) - Any number of features added by camera
manufacturers that affect the video in special ways. Includes fades, wipes,
and solarization.

still video - Recording a single frame of video to several seconds of
videotape, essentially creating a still image that can be annotated with
audio (i.e., use the audio recording tracks to record information about the
picture).

S-VHS (Super VHS) - the same as standard VHS except that the
luminance carrier is shifted to a higher frequency, allowing for greater
carrier bandwidth and, hence, greater resolution (about 400 TVL).

S-VHS -C- A piece of equipment using S-VHS videotape in a smaller
cassette.

synchronization signal - a signal used to synchronize pieces of video
equipment to a common clock. In medium- and large-sized video
facilities, it is necessary to synchronize all pieces of equipment to avoid
problems when recording or playing video.

TIFF (Tagged Image File Format) - a standardized file format used to store
images.

time-lapse - The technique of recording one frame at a time at specified
intervals. When played back at normal speed, time appears compressed,
allowing viewing of a whole day's worth of video in just a few minutes.

tint - See hue.

titling - Referring to the ability to overlay text or symbols onto a video
signal. An example of titling is credits at the beginning or end of a movie.

TVL (television lines) - a unit of horizontal resolution for video devices.

VCR (video cassette recorder) - denotes all formats of video tape recorder
except reel-to-reel.

VHS (video home system) - a piece of equipment using in video tape and
a cassette approximately 4 in by 7 1/2 in.

VHS -C - A piece of equipment using standard VHS video tape in a
smaller cassette.

video - An electrical signal containing timing (synchronization),
luminance (intensity), and often chrominance (color) information that,
when displayed on an appropriate device, gives a visual image or
representation of the original image sequences.

viewfinder compatibility - Implies that a camera or camcorder has a jack
to which an LCD monitor can be attached.

white balance - A camera control that controls the overall intensity of a
video signal. Most cameras come with an automatic white balance
adjustment that can be overridden in situations where the content of the
scene is not "average" (i.e., the subject is either lighter or darker than
average).

wind screen - a device (typically sponge rubber) that is used to cover a
microphone and prevent wind from striking the diaphragm and causing
extraneous (usually annoying) noise while still allowing sound waves to
pass through, creating an audio signal.

wireless playback - A feature on some camcorders and recorders that
allows playback on a television or monitor without physically connecting
wires. This is accomplished through the use of an infrared transmitter in
the camcorder/recorder and an infrared receiver that needs to be attached
to the television/monitor. The receiver is usually included as a part of the
package.

YIQ - Luminance, In-phase, Quadrature (the letter Y is commonly used in
video work as a symbol for luminance).

--------------------------------

Appendix A. Information Resources on the Web

Below is a list of web addresses for companies selling video equipment
that might be useful in video surveillance applications. This is not a
comprehensive list. Inclusion or exclusion neither implies that the
products of one company are better than the products of another for a
given application, nor does it imply that all claims made by these
companies are accurate. Before purchasing any product, check as many
options as possible.

--------------------------------

Appendix B: Effect of Low Light Situations on Cameras

To provide more detailed information on the effects of shooting in low-
light situations, testing on actual video gear was conducted in a major
Federal video-quality laboratory. This appendix illustrates how image
quality changes as light levels change. In addition, one image
enhancement technique, useful for extracting a useable image from
footage taken in too little light, is demonstrated. This technique can be
employed by someone with moderate computer skills working with
commercial software on a standard personal computer.

B.1 Working in Less-Than-Ideal Light

To illustrate the effects of using cameras in low-light situations,
experiments were conducted to show how the ability to distinguish human
faces changes as light decreases. Furthermore, the experimental results
illustrate the impact of both optical and digital zoom and differences in
manufacturers' claims of a certain light level rating.

This test was designed and conducted in this way. A one-eighth scale
photograph of a recognizable person was mounted in front of the camera
in a controlled lighting environment. Each of the two camcorders in the
test were aimed at the picture and positioned such that the head and
shoulders of the individual in the picture filled the frame at each of three
distances: minimum, full optical zoom, and full digital zoom. Lighting
levels started at a level high enough to generate a good quality picture and
were then decreased to approximately 2.4 lux. At that point, a 0.9 neutral
density filter (which blocks 90 percent of the light passing through) was
added to the camera lens, allowing the room lights to be 10 times brighter
than what the camera actually was seeing. Using this technique allowed
the experiment to proceed to the camcorders seeing an effective light level
of 0.1 lux. Light readings were taken with a Tektronix J18 Photometer and
J1811 Illuminance Head with the sensor positioned over the face in the
photograph. Once the reading was taken, the sensor was moved aside, and
video of the photograph was recorded.

Table B-1 lists the camcorders used in this experiment and the relevant
specifications of those devices. One might notice the maximum aperture
varies for Camera B but not for Camera A. This is because Camera B uses
a variable aperture zoom. Variable aperture zooms generally have a
smaller maximum aperture as the lenses are zoomed to their highest
magnification. This is a disadvantage in low-light surveillance situations,
but there are tradeoffs. Variable aperture lenses are smaller and less
expensive to design and manufacture than fixed aperture zooms, such as
the one used in Camera A.

Figures B-1 and B-2 show the facial identification ability of Camera A and
Camera B (respectively) at 14 different light levels. There are many things
to note in these two figures. For Camera A, the minimum light level to
achieve facial identification is about 0.8 lux. For Camera B, it is about 1.5
lux. This is especially interesting given that Camera B has a lower
minimum light rating. In fact, it is useful to compare image quality of the
two camcorders at their respective minimum light rating. Camera A
provides a much more colorful and identifiable face at approximately 4 lux
than Camera B does near its rated 2.5 lux. This is just visual evidence of
the lack of measurement standards for camera and camcorder
specifications. In both sets of images, one can notice "hot spots" or spots
that are brighter than they should be. This is due to reflections off the
image. Because the camera lenses had such a wide field of view for
generating these data, it was possible for them to pick up the brightness of
the lighting source reflecting off the glossy image paper. Finally, for this
part of the experiment, the lens of Camera A was about 9.5 in from the
image, while the lens of Camera B was about 8.5 in distant. The required
difference in placement is consistent with the difference in the shortest
focal length for the lens: 5.9 mm for Camera A and 5.2 mm for the
Camera B.

Figures B-3 and B-4 are similar to B-1 and B-2 except they were taken at
the maximum optical zoom levels of Camera A and Camera B,
respectively. Again, note the minimum useable light level for Camera A is
about 0.8 lux. Camera B, however, does not produce an identifiable image
below 3.4 lux, a significant shift from closest zoom range. This is due to
the variable aperture zoom employed. The shift of minimum acceptable
light from 1.5 lux to 3.4 lux mirrors the change in maximum aperture from
f 1.8 to f 3.2. For this part of the experiment, Camera A was positioned at
54.5 in from the subject while Camera B was positioned at 69.5 in. Again,
the difference is consistent with the focal length of their lenses (59 mm for
Camera A and 72.8 for Camera B).

Figures B-5 and B-6 reveal the effects of using digital zoom in addition to
the optical zoom. Immediately noticeable is the increased speckling or
grain. This is because the cameras only use a portion of the CCD array to
pick up the image: 50 percent for Camera A, 40 percent for Camera B.
Camera B shows significantly more degradation from using the digital
zoom than Camera A does. It also requires significantly more light for a
useable picture. The lowest acceptable light level for identification is 31
lux. For Camera A, it is still possible to identify the subject at 4.3 lux. For
this experiment, Camera A was 131 in from the subject and Camera B was
175 in.

B.1.1 Enhancing the Images

There may come an occasion when it is absolutely necessary to record
video in light levels below what is known to be acceptable for the
purposes of the surveillance. In these situations, it is best to take the
highest quality video possible (i.e., stable camera, as little motion in the
scene as possible). Afterwards, it may be possible to extract some useful
information from the tape using image processing techniques.

One of the simplest techniques involves "capturing" or recording a
sequence of video frames to a computer's hard disk and then averaging the
images to improve the signal-to-noise ratio. Figure B-7 shows the effects
of averaging 30 frames of video for Camera A at maximum optical zoom.
Note it is possible to identify the subject at a light level of 0.23 lux.[9]
This is a significant improvement over the 0.8-lux light level that was
required without averaging (fig. B-3). The procedure to accomplish this
follows.

To begin, the video was taken using Camera A, a digital camcorder. The
camcorder was connected to an IBM-compatible personal computer (133
MHz Pentium [trademark] running Microsoft Windows 95 [registered
trademark]) with the Canon Video DK-1 DV Capture Kit installed using
the cable supplied with the capture kit. (This capture kit is compatible with
all digital camcorders that have an IEEE 1394 "Fire Wire" interface, and
has no relationship to the manufacturer of Camera A or Camera B.) Using
the supplied software (DV Commander [registered trademark]), 30
consecutive frames were saved on the PC's hard drive. The frames had 640
x 480 pixel resolution and were saved as TIFF (Tagged Image File
Format) images. The files were then individually loaded into a scientific
computing package (IDL [registered trademark]), where they were
individually scaled to maximize contrast. All the images were summed on
a pixel-by-pixel basis, and then that result was divided (also pixel-by-
pixel) by the number of frames (30). The resulting image was once again
scaled to maximize contrast and then saved to a separate file. Each image
in figure B-7 is cropped[10] from one of these averaged files.

While image processing using IDL [registered trademark] might be
beyond the average computer user, there are PC-based graphics-design
software packages that can be used to achieve the same goal. One such
package is Adobe [registered trademark] Photoshop [registered
trademark]. Using Photoshop [registered trademark], one can read the
TIFF images as separate files. For each image, the content can be copied
and pasted into a layer of a master image. Once in a master image, each
layer should be adjusted using the "Auto Levels" feature (pressing
Control+Shift+L simultaneously), and the opacity should be adjusted to
100/number of frames percent. (For this reason it is best to try to have a
number of frames that will evenly divide into 100.) Once all layers have
been adjusted, the layers can be combined (i.e., flattening the image) and
the image can be saved.

B.2 Summary

This appendix has provided an overview of how diminishing light can
effect the images produced by video equipment. In doing this, it showed
how the low-light threshold of a camera or camcorder could be visually
assessed. Finally, a brief introduction into image enhancement was given
showing useful information can be extracted from videotape footage even
when direct viewing does not reveal anything useful.

--------------------------------

Endnotes

1. Certain commercial companies, equipment, instruments, materials, and
organizations are mentioned in this report to adequately explain the
experiments and their results. In no case does such identification imply
recommendation or endorsement by the National Institute of Justice, or
any other U.S. Government department or agency, nor does it imply that
those identified are necessarily the best available for the purpose.

2. Section 4 will explain "lux" and present examples of different lux
levels.

3. NIJ Guide 200-98 and other NIJ guides can be ordered from
NIST/OLES , 100 Bureau Dr., Stop 8102, Gaithersburg, Maryland, 20899-
8102.

4. The resolution of NTSC video equipment is measured by the number of
vertical lines that can be distinguished (horizontally) across a frame of
video. This is because (1) vertical resolution is fixed and (2) one gets an
indication of how much information the frame contains, regardless of the
size of the input or viewing device.

5. It is possible to have situations where desired signals for some people
become the noise source for other people and their systems.

6. If the operator of video surveillance equipment has some control over
the environment where the surveillance will take place, some improvement
in SNR may be possible. That is, an increase in illumination (light level)
will increase SNR (as explained in the next section).

7. It is true that "wide angle" and "telephoto" lenses can be mounted in
front of the camcorder's permanently mounted lens, increasing the range of
situations in which the device can be used. These lenses do, however,
generally reduce the amount of light reaching the lens by one or two f-
stops.

8. For more information on batteries, consult the "New Technology
Batteries Guide," NIJ Guide 200-98. This and other NIJ guides are
available from NIST/OLES, 100 Bureau Dr., Stop 8102, Gaithersburg,
Maryland, 20899-8102.

9. At a light level of 0.23 lux, the image in the viewfinder of the
camcorder was almost totally noise. Only the vaguest outline of the person
in the image was discernable.

10. The cropping was done only to provide a more compact display in the
figure.

Order your surveillance system 

---

Security and Safety Supply, police products, police supply