Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 2 Current »

Technical Article | TA-20201002-TP-14

VDG Sense | Video Content Analyses | ObjectR Basic

PDF Download

Note: ObjectR Basic is not available in VDG Sense 2.5 or higher!

Pre-defined rules isolate and classify objects directly from the current video stream. Activation events automatically in case of violation of defined rules. This function allows you to define the detection of left, stay in a designated area of the determination of the authorized period of time. The event is correlated with the activation of the macro system rapidly releasing further, related scenarios including notification system operator alarm.

Abandoned object detection is performed based on the detection of a situation in a zone of the moving object has become the object stationary or separated from another object in the specified during the setup period of time.

Detecting emerging object takes place on the basis of the detection of the situation in the zone where there is a stationary object which has not appeared before in any point of the scene.

This technique can only be used in quiet areas in which it is not normal for something to be left behind. It cannot be used in crowded areas.

Requirements of camera mounting spot to ensure the highest level of efficiency

  • Apply in quiet areas

  • Appropriate choice of abandoned object detection time

  • Mounting the camera, which provides a view of the entire width of the platform, prevent obscurance  by other persons

  • Mounting the camera high enough to avoid obscuring the crowd that each person was seen as a separate object’

  • Desired lighting scene should be more than 5 lux

 

Recommended image from the camera (quit areas):

Limitations of the algorithm

  • Big crowds or much movement in video will limit effectiveness

  • Obscurance of the left of the object by another object, eg. a moving person covers luggage, etc.

  • Too short detection time could generate a large number of false alarms

  • Too long detection time could delay or prevent objects triggering an alarm. When the countdown time to leave the object is interrupted by covering the next person that appears

  • The inability to distinguish between the abandoned luggage from baggage located by a person

  • Low contrast of the object in relation to the environment

  • No labels