Persistent Change Detection

Detects moving objects and generates an alert when an object becomes stationary and remains stationary, causing a change in the scene. You can use this engine to detect dropped or abandoned bags. This engine requires video input (images are not supported).

Configuration Parameter Description
FreeformRegion The region of the scene that you want to monitor.
Input The name of the image track to analyze.
MaxSize The maximum expected size of an object.
MinimumTime The minimum amount of time that an object must remain stationary.
MinSize The minimum expected size of an object.
NumParallel The maximum number of video frames to analyze simultaneously.
Region The region of the scene that you want to monitor.
SampleInterval The interval at which frames are selected to be analyzed.
Type The analysis engine to use. Set this parameter to PersistentChangeDetection.

Output Tracks

Output track Description
Data Contains one record for each object that is stationary for longer than MinimumTime, for each video frame.
Result Contains one record for each object that is stationary for longer than MinimumTime.
ResultWithSource

The same as the Result track, but each record also includes the best source frame.

Start

The same as the Data track, except it contains only the first record of each event.

End

The same as the Data track, except it contains only the last record of each event.

Alert

Contains one record for each object that is stationary for longer than MinimumTime.

This track differs from the Result track because records are created as soon as the object meets the minimum time requirement, rather than when the object leaves the scene.

AlertWithSource

The same as the Alert track, but each record also includes the source frame.

To see an example record, use the action GetExampleRecord.