|
The subject of specifying cameras is a jungle of jargon and
misinformation, this brief article attempts to shed a little light on
some of the mysteries surrounding it. Only CCD cameras will be
considered because they are now the most commonly used type for CCTV.
CCD means a Charged Coupled Device and consists of a flat array of
tiny, light sensitive photodiodes. Each diode produces a voltage that
is directly proportional to the amount of light falling on it. No light
would produce no voltage and therefore a black level. Maximum light
would produce a maximum voltage and therefore a white level. In between
these would be shades of grey, and is the luminance information of a
video signal. In the case of a colour camera, a chrominance signal is
superimposed onto the luminance signal to carry the colour information.
(If a colour camera is connected to a monochrome monitor, then a
monochrome picture would be produced from the luminance information and
the chrominance would not be processed). See also colour cameras with
separate Y/C outputs under resolution.
Image:Constant60.gif
The range of light levels that a CCD can cope with is very
limited, therefore means have to be introduced to restrict the light
range within certain limits.
A field of video is created by the CCD being scanned across and down
exactly 312 1/2 times and this reproduced on the monitor. A second scan
of 312 1/2 lines is exactly 1/2 a line down and interlaced with the
first scan to form a picture with 625 lines. This is known as a 2:1
interlaced picture. The combined 625 line is known as a frame of video
and made up from two interlaced fields. The total voltage produced is
one volt from the bottom of the sync pulse to the top of the white
level, hence one volt peak to peak(p/p). The luminance element of the
signal is from 0.3 volts to one volt, therefore is 0.7 volts maximum.
This is known as a composite video signal because the synchronising and
video information are combined into a single signal.
Image:Constant61.gif
Note that the imaging device is scanned 625 times but the actual
resolution is defined by the number of pixels making up the device.
There are several factors that make up a complete camera specification and are all be inter-related. These are:
Sensitivity
Signal to noise ratio.
Automatic gain control.
Resolution.
The most common factor people look for in a camera specification is
the sensitivity, although it is not always the most important.
Sensitivity is the amount of light, in lux, necessary to produce a
video signal of some, usually unspecified, level. This factor seems to
be the marketing battleground upon which all manufacturers fight to
show their cameras as being better than the competition!
As seems obvious this is the ratio of the level of the video signal
to the amount of noise present. Noise in a video is seen as snow or
graininess, resulting in a poorly defined image on the monitor or video
recording. The unit for expressing s/n ratio is decibels (dB), but do
not be too worried because it can be expressed as a ratio. The
following table shows the equivalent ratio for values given in dB.
| dB |
Ratio |
| 100 |
100,000:1 |
| 60 |
1,000:1 |
| 50 |
316:1 |
| 40 |
100:1 |
| 30 |
32:1 |
| 20 |
10:1 |
| 10 |
3:1 |
It can be seen that a s/n ratio of 40Db is equivalent to a ratio of
100:1, that is the signal is 100 times the noise level. Conversely the
noise is one hundredth of the signal. Note that at a s/n ratio of 20Db,
the noise is 10% of the signal and would produce an unacceptable
picture. The following table provides a guide as what quality to expect
from various s/n ratios.
| S/N ratio dB |
S/N ratio:1 |
Picture quality |
| 60 dB |
1,000 |
Excellent, no noise apparent |
| 50 dB |
316 |
Good, a small amount of noise but picture quality good. |
| 40dB |
100 |
Reasonable, fine grain or snow in the picture, fine detail lost. |
| 30 dB |
32 |
Poor picture with a great deal of noise. |
| 20 dB |
10 |
Unusable picture. |
When the light falling on to an imaging device reduces to a certain
level, there is insufficient to create a full level video signal. AGC
acts to increase the amount of amplification in these conditions to
bring the signal up to the required level. As well as amplifying the
video signal, additional noise can be introduced, and the signal to
noise ratio reduced. The result is frequently a very much degraded
signal and poor picture on the monitor.
The value referred to here is the horizontal resolution in TV lines,
that is the number of black to white transitions that can be resolved
across the image. This is a function of the number of pixels that make
up the CCD imaging area and the bandwidth of the camera circuitry.
Typical camera resolution is 350 TV lines, with high resolution cameras
producing better than 450 lines. Note that resolution costs money!
There are now colour cameras that instead of superimposing the
chrominance onto the luminance signal, provide the chrominance as a
separate signal. This is known as Y/C separation and requires two
coaxial cables from the camera to carry each signal separately. The
effect of this technique is to increase the bandwidth and therefore the
resolution, typically to better than 500 TV lines.
The human eye is an incredibly adaptable device that can focus on
distant objects and immediately re-focus on something close by. It can
look into the distance or at a wide angle nearby. It can see in bright
light or at dusk adjusting automatically as it does so. It also has a
long 'depth of field' therefore scenes over a long distance can be in
focus at the same time. It sees colour when there is sufficient light
but switches to monochrome vision when there is not. It is also
connected to a brain that has a faster updating and retentive memory
than any computer. Therefore the eyes can swivel from side to side and
up and down, retaining a clear picture of what was scanned. The brain
accepts all the data and makes an immediate decision to move to a
particular image of interest. It can then select the appropriate angle
of view and re-focus. The eye has another clever trick in that it can
view a scene of great contrast and adjust only to the part of it that
is of interest.
By contrast the basic lens of a CCTV camera is an exceptionally
crude device. It can only be focused on a single plane, everything
before and after this becomes progressively out of focus. The angle of
view is fixed at any one time it can only view a specific area that
must be predetermined. The iris opening is fixed for a particular scene
and is only responsive to global changes in light levels. Even an
automatic iris lens can only be set for the overall light level
although there are compensations for different contrasts within a
scene. Another problem is that a lens may be set to see into specific
areas of interest when there is a lot of contrast between these and the
surrounding areas. However as the sun and seasons change so do light
areas become dark and dark areas become light so the important scene
can be 'whited out' or too dark to be of any use.
One of the most controversial but important aspects of
designing a successful CCTV system is the correct selection of lens.
The problem is that the customer may have a totally different
perspective of what a lens can see compared to the reality. This is
because most people perceive what they want to view as they see through
their own eyes. Topics such as identification of miscreants or number
plates must be subjects debated frequently between installing companies
and customers.
The selection of the most appropriate lens for each camera
must frequently be a compromise between the absolute requirements of
the user and the practical use of the system. It is just not possible
to see the whole of a large loading bay and read all the vehicle number
plates. The solution may be more cameras or viewing just a restricted
area of particular interest. The company putting forward the system
proposal should have no hesitation of pointing out the restrictions
that may be incurred according to the combination of lens versus the
number of cameras. Better this than an unhappy customer who is
reluctant to pay the invoice.
These are sometimes referred to as monofocal lens. As the name
implies this type of lens is specified when the precise field of view
is fixed and will not need to be varied when using the system. The
angle of view can be obtained from the supplier's specification or
charts provided. They are generally available in focal lengths from
3.7mm to 75mm. Longer focal lengths may be produced by adding a 2x
adapter between the lens and the camera. It should be noted that this
will increase the f number by a factor of two (reducing the amount of
light reaching the camera). If focal lengths longer than these are
required then it will be necessary to use a zoom lens and set it
accordingly.
Except for very wide angle lenses all other lenses have a ring
for adjusting the focus. In addition cameras include a focusing
adjustment that moves the imaging device mechanically relative to the
lens position. This is to allow for minor variations in the back focal
length of lens and manufacturing tolerances in assembling the device in
the camera. Correct focusing requires setting of both these
adjustments. The procedure is to decide the plane of the scene on which
the best focus is required and then set the lens focusing ring to the
mid position. Then set the camera mechanical adjustment for maximum
clarity. Final fine focusing can be carried out using the lens ring.
The mechanical focusing on cameras is often referred to as the
back focus. This was because a screw at the back of the camera moved
the tube on a rack mechanism. Modern cameras now have many forms of
mechanical adjustment. Some have screws on the side or the top, some
still at the back. There are cameras that have a combined C/CS-mount on
the front that also has the mechanical adjustment and can accept either
type of lens format. The longer the focal length of the lens the more
critical is the focusing. This is a function of depth of field.
This is a design of lens that has a limited range of manual focal
length adjustment. It is strictly not a zoom lens because it has quite
a short focal length. They are usually used in internal situations
where a more precise adjustment of the scene in view is required which
may fall between two standard lenses. They are also useful where for a
small extra cost one lens may be specified for all the cameras in a
system. This saves a lot of installation time and the cost of return
visits to change lenses if the views are not quite right. For companies
involved in many small to medium sized internal installations such as
retail shops and offices this can save on stockholding. It makes the
standardisation of systems and costing much easier.
A zoom lens is one in which the focal length can be varied manually
over a range by means of a knurled ring on the lens body. It has the
connotation of 'zooming in' and therefore infers a lens with a longer
than normal focal length. The zoom ratio is stated as being for
instance 6:1 this means that the longest focal length is six times that
of the shortest. The usual way of describing a zoom lens is by the
format size, zoom ratio and the shortest and longest focal lengths,
i.e. 2/3," 6:1, 12.5mm to 75mm. Again, great care must be taken in
establishing both the camera and the lens format. The lens just
described would have those focal lengths on a 2/3" camera but a range
of 8mm to 48mm on a 1/2" camera. Similarly a lens giving the same
performance on a 1/2" camera would be a 1/2," 6:1, 8mm to 48mm.
Manual zoom lenses are not widely used in CCTV systems because the
angle of tilt of the camera often needs to be changed as the lens is
zoomed in and out. The most common need for a zoom lens is when used
with a pan tilt unit. The lens zoom ring is driven by tiny DC motors
and controlled from a remote source. With a correctly set up camera
lens combination the focus should not change from one limit of zoom to
the other.
With the development of ever smaller cameras and longer focal
length lenses the method of mounting the camera/lens combination must
be taken into account. There are many cases where the lens is
considerably larger than the camera and it may be necessary to mount
the lens rigidly with the camera supported by it. In other cases it may
be necessary to provide rigid supports for both camera and the lens.
Always check the relationship between the camera and lens sizes and
weights when selecting a housing or mounting. Most manufacturers of
housings can provide lens supports as an accessory.
The most frequent reason for the focus changing when zooming is
that the mechanical focus of the camera has not been set correctly.
There are many situations where it is required to pan tilt and zoom
to a predetermined position within the area being covered. It is
possible to obtain motorised lenses with potentiometers fitted to the
zoom and focusing mechanisms. These cause the lens to zoom
automatically and focus to the setting by measuring the voltage across
the potentiometer and comparing it with the signals in the control
system. All other functions are as for motorised zoom lenses. Pre-set
controls are only possible with telemetry controlled systems. The
specification of the telemetry controls should be checked to see
whether the pre-set positions are set from the central controller or
locally from the telemetry receiver.
|
