Video requirements to be met for abandoned object detection tool of the Tracker object operation

1. Camera requirements:
1. Resolution must be not less than 640x480 pixels.
2. Analytics works both with color and monochromatic images.
3. It is not allowed to the image shift be more than 1% from the frame size due to camera shake.
2. Lighting requirements:
1. Moderate lighting. The quality of analytics performance can be lower in case of low lighting (night) or over lighting (overexposure).
2. There must be no dramatic changes in lighting.
3. Scene and camera angle requirements:
1. Background is mostly static and is not changed.
2. Incorrect analytics performance can be caused by reflective surfaces.
3. Abandoned objects are hidden by moving objects not longer than 10% of the time.
4. Abandoned object images requirements:
1. An abandoned object must be visible on the image.
2. There must be no noise on the video image and there must be no artifacts caused by the compression algorithm.
3. Object dimensions on the video image must not be less than 1% of the frame size.
4. The width and height of objects in the video should not be less than 1% of the frame size (at resolutions not less than 1920x1080 pixels) or 15 pixels (at lower resolutions).
5. The width and height of objects in the video should not be more than 25% of the frame size.

If these requirements are met, the following performance indicators of the abandoned object detection tool are guaranteed:

• 92 of 100 abandoned objects are detected.
• Out of 100 detection tool triggerings, 80 will be caused by real abandoned objects, and 20 – by other changes in the scene under surveillance.