Features of installing TV cameras for intelligent video systems.

Currently, we are witnessing a general transition from traditional TV systems to network ones. This entails the use of various IP video codecs, including those with built-in video analytics. As a result, the video surveillance system becomes intelligent — it is able to independently interpret the situation in the field of view, make decisions and provide the user with only the necessary information. But the «intelligence» of the system can be reduced to zero by non-compliance with a number of simple rules that lie on the surface and are often ignored. This article provides information on the basic principles of installing and configuring stationary and PTZ cameras at a facility designed to operate as part of an intelligent video system (IVS). It accumulates the experience of our company, as well as our partners — IOIMAGE (Israel), in installing video analytics devices at various facilities. Since each facility may have specific safety requirements and these requirements may vary depending on specific conditions, the information below is basic and subject to creative revision in each specific case.

Let's consider a situation when the system is built on the basis of traditional equipment and IP encoders with built-in video analytics. It is the most common nowadays — not all TV equipment installed in Russia is obsolete and requires replacement. In this case, the following general principles for connecting stationary and PTZ cameras to video signal encoders must be observed:

If the TV camera has a built-in character generator (Onscreen Display), then it must be turned off.

? PTZ cameras must have autofocus. It is necessary to make sure that autofocus (AF) is on, and manual focus adjustment (MF) is turned off.

? Backlight compensation (BLC) must be disabled for PTZ cameras.

? The camera should be positioned in such a way as to exclude direct sunlight from entering the lens at any time of the day.

? If the TV camera has a built-in image stabilizer, it must be activated.

? If the TV camera has a built-in motion detector, it must be disabled.

? If the PTZ camera provides masking of certain areas in the field of view, this function must be disabled.

? If the TV camera has a «night» mode with an increased accumulation time, this function must be disabled.

? A TV camera should give a clear idea of the object that poses a potential threat, and not create a pleasing picture for the eye. All camera settings (gain, clarity, gamma correction, etc.) should be subordinated to this principle.

? Cameras should be focused on the object as follows:

– For all cameras, the focus is initially set to ensure maximum depth of field during the daytime, final fine-tuning is done at night.

– For dual-channel day/night cameras, each channel is configured independently of each other.

Lens requirements:

– Most standard lenses have an IR cut-off filter. If the facility plans to use IR illumination at night, the lens must ensure that the IR filter is turned off at night.

– It is necessary to use a lens with a magnification sufficient to identify the intruder in the required range. Remember that the higher the magnification, the smaller the field of view.

? TV camera installation height and field of view:

– To create the best detection conditions when an object is moving towards the camera, the installation angle (the angle between the viewing axis and the horizon line) and the installation height should be selected based on the size of the detection zone and the possible directions of the intruder’s movement. In general, a higher installation height and a lower installation angle provide the best visualization of movement in the detection plane when the intruder is moving towards or away from the TV camera. When the camera’s viewing axis is directed almost parallel to the horizontal surface (the installation angle is close to 90°) and the camera is installed at a low height, the visualization of movement will be less pronounced. In this case, the object will visually change in size with minor movement, but it is practically impossible to calculate its element-by-element movement in the horizontal plane. Wide-angle lenses provide smaller sizes of the detection zone due to the fact that at a large distance the image of the intruder degenerates into a point.

osobennosti ustanovki tv

– Figure 1 explains how the camera installation angle affects the detection process. The closer the intruder is to the camera, the more acute the angle at which he is observed. Accordingly, the greater the number of image elements per step when moving towards or away from the camera. If the camera installation angle is 90°, the object visually changes in size when moving, but in the detection plane (horizontal surface) the element-by-element shift cannot be calculated.

Fig. 2. Optimizing the overlap of fields of view

Lighting considerations:

– The light source should always be located behind the camera, not in front of it. If the light source is located in front of the camera, it may cause the lens's auto iris to operate in such a way that its newly set value will be insufficient for low-light areas of the field of view.

– Lighting (including infrared) should not create overexposure on the underlying surface in the near field of view of the TV camera. This is also necessary to prevent incorrect operation of lenses with auto iris.

– When using visible light sources, the illumination of the horizontal surface at night, measured by a lux meter, should be at least 5 lux – video analytics devices work effectively only with a high signal-to-noise ratio.

? Mutual arrangement of cameras on the object. If the system uses PTZ camera guidance based on target designation signals from fixed cameras, the PTZ camera must compare the three-dimensional coordinates of the object in the fixed camera's field of view with its own presets. Errors in measuring coordinates depending on the angle of the object and the field of view of the lens were considered above. In a situation where the error in measuring the range coordinate by a fixed camera is large, this can lead to undesirable consequences. If the viewing axes of the fixed and PTZ cameras are initially located at an angle of 90°, the total sensitivity of the system to the error in measuring coordinates becomes more noticeable, since an error of several elements leads to an error of several meters horizontally for the PTZ camera. When designing the overlap of the fields of view of PTZ and fixed cameras, the following recommendations can be used to understand the accuracy of PTZ camera positioning at different installation angles of the fixed camera:

– For a fixed camera with a small installation angle and a narrow vertical field of view, the following figure shows the optimization of the overlap of the fields of view between the PTZ camera and the fixed camera (Fig. 2A).

– For a fixed camera with an average installation angle or average vertical field of view (Fig. 2B), it shows different quality of optimization of the total field of view.

– For a fixed camera with a large installation angle or a wide field of view in the vertical plane (Fig. 2B) illustrates the quality of the video surveillance system installation. – In cases where the surveillance area is of significant size, cameras should be positioned in such a way as to eliminate the appearance of “blind spots”. The field of view of the cameras should be calculated in such a way that in any case it is possible to observe the intruder in full growth.
– Cameras should be positioned so that any attempt to disable one camera would be recorded by another camera.
– The camera installation location should be chosen as a compromise between the required field of view and considerations of equipment safety, ease of installation, and access during maintenance.

? PTZ cameras:
– If the PTZ camera supports digital zoom, it must be disabled. Only optical zoom should be used in the temporary detention facility.
– If the camera has a zoom speed setting, its optimal speed must be agreed upon with the video analytics provider.
– It is necessary to disable all controllers and keyboards for controlling PTZ cameras, except for those that are directly used in the temporary detention facility.

Using additional keyboards may lead to undesirable effects in the camera control circuit and collisions between decisions of the temporary detention facility operators.
For all cameras without exception, it is necessary to draw up a maintenance schedule, including cleaning the porthole of the hermetic housing.

Fig. 3. Incorrect placement of cameras on site

Fig. 4. Correct placement of cameras on the site

In conclusion, we note that only strict adherence to these basic rules will justify the costs of purchasing expensive IVS equipment and obtain specific results of its implementation at the facility.

______________________________

V. Nikitin
Magistral’SKN group of companies

Source: Security Algorithm magazine №6, 2010