VideoIntellect abandoned objects detection

The VideoIntellect abandoned objects detection is configured as follows:

1. On the Select area tab, do the following:
   
   1. Set one or more zones in which the abandoned objects will be detected (1).
   2. To capture a frame of a video image, click the Stop video button (2).
   3. All set zones are displayed in the Detection zones area (3). If necessary, you can change the name of the zone or delete the zone.
   4. Click the Apply configuration changes button (4).
2. On the Abandoned objects detection tab, configure the detection tool parameters:
   
   1. In the Initial detection sensitivity (0-1) field (1), enter the general sensitivity of the detector. Value range is from 0 to 1. Value 0 means that no object will be detected, value 1 means maximum sensitivity. The recommended value is 0.7.
   2. In the Object visibility (0-1) field (2), enter the detection sensitivity of subtle (merging with the background) objects. This parameter indicates the ability of the detector to distinguish subtle (merging with the background) objects.
      Value range is from 0 to 1 in conventional units. Value 0 means the object will not be detected, value 1 means maximum sensitivity. The recommended value is 0.95. The recommended value range is from 0.82 to 0.97.
   3. In the The passenger traffic density (0-1) field (3), enter the object detection sensitivity in conditions of intensive object overlap by passing people.
   4. In the Minimum size of detected object (cm) field (4), enter the minimum object size on one side in centimeters to be detected. Smaller objects will not trigger the detection. The size of the object is specified relative to the frame.
   5. In the Maximum size of detected object (cm) field (5), enter the maximum object size on one side in centimeters to be detected. Larger objects will not trigger the detection. The size of the object is specified relative to the frame.
   6. In the Detection time (s) field (6), enter the time in seconds after which the object is considered abandoned.
   7. In the Shadow/highlight filter type drop-down list (7), enable additional filtering of shadows and highlights (disabled by default). Select one of the filter types: "z003", "z002", "z001", "zac", "zu", "zz" depending on the scene contrast, lighting variability, presence of shadows and highlights. For outdoor areas with shadows and visual noise (caused by grass and other plants, for example), and for objects with glares, the filter types "z003" and "zz" are recommended. For underbridge spaces with shadows and contrasting objects, the filter types are “z003” and “zac” are recommended. The filter is selected empirically.
   8. Set the Without person checkbox (8) to ignore the objects left next to the person.
   9. Set the Filter objects checkbox (9) to enable the object neural network filter. 
3. On the Parameters tab, the area (1) displays the current detection configuration. This configuration can be copied to any text file. To upload a configuration from a file, click the Load configuration from file button (2).
   
   Description of configuration parameters:
   1. sensitivity—Initial detection sensitivity.
   2. rejectlvl—Object visibility.
   3. acceptlvl—The passenger traffic density.
   4. objsizemin—Minimum size of detected object.
   5. objsizemax—Maximum size of detected object.
   6. detectiontime—Detection time.
   7. isFilterHumans—Without person.
   8. isFilterObjects—Filter objects.
4. Click the OK button to complete the detection configuration.

The VideoIntellect abandoned objects detection is now configured.