Documentation for Intellect 4.10.4. Documentation for other versions of Intellect is available too.

Previous page Creating and configuring VMDA metadata storage  Creating and configuring the Tracker object Next page


Before configuring the smart video detection tools (see Creating and configuring the Tracker object), consider the following recommendations that could help to reduce the false alarms:

  1. Configure the Tracker object iteratively and check its operation quality at each iteration.
  2. The parameters that affect the Tracker object operation quality the most are: Minimum size, Maximum size, and Sensitivity.
  3. It is recommended to set Sensitivity to zero (left end position of the slider). This will enable automatic sensitivity ranging to ensure reasonable operation quality of the tracker in changing lighting conditions.
  4. If lighting is stable, it could be reasonable to adjust the Sensitivity value manually. To detect the low contrast objects, set the Sensitivity value at about 35, for the high contrast ones — at about 15.
  5. Minimum size is to be selected so that it is a little less than typical object size on the image.
  6. Maximum size is to be selected so that it is a little greater than typical object size on the image, considering that the object can be joined with its shadow.
  7. Lowering the Waiting for loss parameter allows excluding triggering of false static objects, while setting it to a higher value could allow not breaking off alarms for overlapping or temporarily disappearing objects in some cases.
  8. To increase the VMDA detection tools accuracy, use an advanced object classifier, which is configured using the scripts and registry keys — see also CAM_VMDA_DETECTOR section of the Programming Guide.
  9. To increase the VMDA detection tools accuracy, the neural filter can be used. It is trained individually for each use case and provides the high accuracy of the objects classification — up to 99% (see Configuring the neural filter).
  • No labels