Back to previous page.
GUIDLINES FOR IDENTIFICATION
Good System design is vital when contemplating installing a CCTV network. Better results will be obtained if the CCTV equipment is not only correctly selected, installed, adjusted and maintained but also designed in accordance with the demands and requirements of the application.
A key consideration is the purpose for which a given lens and camera combination that will provide in information that is required. As a result, selecting the correct capturing equipment, including the lens, is critical the system’s success.
In most applications the primary goal of the system may be reduced to three basic requirements; Detection, Recognition and Identification.
Since detection, identification, and recognition are all visual processes, a figure must be detected, and then recognized at the farthest range possible, to record the event. As a result, these 3 standard viewing levels of the figure have been used by CCTV system designers to determine the proper lens to be used.
Viewing Levels Definition:
☑ Detection Standard: The captured image should occupy not less than 10% of the screen size.
☑ Recognition Standard: The captured image should occupy not less that 50% of the screen size.
☑ Identification Standard: The captured image should occupy not less than 120% of the screen size.
The value for various degrees of identification are given as the percentage of human figure with an ‘average’ height of 1.6M would occupy the monitor screen. This is called “Screen Height Ratio”. Refer to Figure 1
Many system engineers stumble when trying to find the right camera and lens combination. The chart ("Horizontal & Angular Fields of View") shows the horizontal and vertical fields of view for many lenses and formats, however it may not be sufficient in determining the right combination of camera and lens for special requirement.
Top of Page
The field of view is the ratio of the sensor size to the focal length and the distance to the subject. The ‘width to height’ ratio of the sensor is 4:3. Refer to Figure 2
The horizontal and vertical angles and therefore fields of view are different and must be considered separately.
The diagram below (figure 3) shows the sensor sizes to be used when calculating fields of view and angles of view.
Supposing it required to recognize a known person at 50M, using a 1/3” CCD formated camera.
Let us say that the required Viewing level is recognition. Therefore the man has occupy 50% of the total view height.
So if the total view height = H = 2 X 1.6m = 3.2m = 3,200mm
And the distance from to the man figure d = 50m = 50,000mm
And the hight of 1/3” CCD h = 3.6mm. Therefore:
Therefore, in this example, an f = 56.25 mm lens would be required to recognize a known man on 50 meter distance,
All that need to be done now is to select a lens which it has the nearest standard focal length.
Top of Page
THE 10% RULE
There is an empirical rule used widely in CCTV system design that in order to distinguish a person on a monitor, they must be at least 10% of the height of the screen. ‘To distinguish’ in this context, means that the image can be seen to be a person. This 10% is also considered to be the minimum resolution necessary for an efficient VMD system to recognise an alarm situation. Many VMD systems will certainly respond to smaller targets, but with the probability of generating a higher number of false alarms, and the difficulty of identifying the cause. There may be occasions where a different factor is required, in these cases the equations can be adapted to the need.
The target size:
The standard target size has been taken as being 1.6 metres high. This represents an average height person passing through the field of view. If a different target such as a person crawling, is a design parameter then the equations can be adjusted accordingly.