The following table contains the requirements for cameras used by the Queue detection tool:
Camera: | - resolution: Resolution: 720 х 576 (CIF4), 360 х 288 (CIF1) to 720 х 576 (CIF4) pixels; lager images are scaled down to CIF4;is also allowed to use. Increasing the resolution above CIF4 does not improve the operating quality of the recognition algorithm
- Frames frames per second: 6 or more;
- colorColor: color or greyscale;
- no No camera jitter is allowed.
|
Illumination: | - best Best recognition results are achieved under moderate illumination. If the scene is under- or over-illuminated, the recognition accuracy may drop down;
- sharp Sharp changes in illumination may lead to improper operation of analytics.
|
Scene and viewing camera angle: | - vertically Vertically downward position of the camera is the best for the purpose. The closer to vertical, the more accurate counting;the estimation
- camera Camera FOV dimensions: min. 3 x 3m (6 x 6 minimum 3x3 m (6x6 humans), optimal 4 x 4m (8 x 8x 4x4 m (8x8 humans), max. 8 x 8m (16 x 16 maximum 8x8 m (16x16 humans);
- the The background must should be primarily static and should not undergo sudden changes;
- reflective Reflective surfaces and harsh shadows from moving objects can affect the quality of analytics;
- Analytics may not work correctly if there are periodic movements of the background objects in the camera FOV (leafage, TV screens or any periodic object movement in the background may cause analytics glitches., etc.)
|
Images of objects within the scene: | - image Image quality: the image must should be clear and sharp , with no visible compression artifacts;
- dimensions Dimensions of a human in scene: bounding rectangle has to occupy from 0.25 25% to 10 percent 10% of the frame area.
|
