For video analytics to work correctly, the following requirements must be met:
- Camera requirements:
- Resolution:
Min. 640x480 pixels for Scene Analytics;
Min. 320x240 pixels for other detection tools
- Frames per second:
Min. 6 fps for Scene Analytics;
Min. 1 fps for other detection tools.
- Color: video analytics work with both black-and-white and color images.
- Camera shaking must not cause image shifting of more than 1% of the frame size.
- Resolution:
- Lighting requirements:
- Moderate lighting. Lighting that is too little (night) or too much (bright sunlight) may impact the quality of video analytics.
- No major fluctuations in lighting levels.
- Scene and camera angle requirements:
- Moving objects must be visually separable from each other in the video.
- The background must be primarily static and not undergo sudden changes.
- Minimal obscuration of moving objects by static objects (columns, trees, etc.).
- Reflective surfaces and harsh shadows from moving objects can affect the quality of analytics.
- Long single-color objects may not be tracked properly.
- Object requirements:
- There is no noise on the video image and there are no artifacts caused by the compression algorithm.
The width and height of the objects in the image must be not less than 1% of the frame size (if resolution is over 1920 pixels) or 15 pixels for lower resolution.
The width and height of the objects in the image must not exceed 75% of the frame size.
- The speed of objects in the frame must be at least 1 pixel per second.
In order to detect the object it is to be visible at not less than 8 frames.
Within two adjacent frames the object cannot move in the movement direction for the distance that is longer than its size. This condition is essential for correct calculation of the object’s trajectory (track).