Estimation of possibility of person detection, person identification and reading a license plate only by the pixel density is convenient, but inaccurate.

The license plate or face of a person can simply not get into the view area in height, or get at too big angle and the task of identification or reading license plate will not be solved.

At the same time, to detect a person it is not necessary to get the whole person in the frame, it is enough to see a part of his body. Possible height range of appearance of a License plate does not equal the height range of appearance of a person’s face.

 

Therefore, the real area of person identification will be less than the colored region on the view area projection, the real area of license plate reading will be shifted relative to the colored region, and the real area of person detection will be more than on the projection.

 

To eliminate this error VideoCAD offers more rigorous tools, taking into account the height restriction and limiting angles. You can read about them by the links below.

 

 

 

 

You can use these tools when you need a more accurate result.

 

See also: "The principles of CCTV design in VideoCAD. Part 2. Person detection area, person identification area, license plate reading area. Pixel density."(*.pdf) .