Do You See What I See?: Whenever You Need Reliable Video Motion Detection, Advanced Algorithms Can Do the Hard Work For You

This article describes the new algorithms being used in conjunction with closed circuit television (CCTV) and other image data capture systems that allow them to determine the content of the data being acquired. Such algorithms are part of video content analysis (VCA), which follows visual rules for what patterns of data may need to be red flagged. Such algorithms do not just keep track of motion, but can also identify things like abandoned luggage, or even behavior deemed suspicious, like loitering. One such set of rules described is the Intelligent Video Motion Detection (IVMD) algorithm. IVMD uses a background learning algorithm so it can exclude unwanted alarm triggers like moving branches, leaves, clouds, shadows, and the like. This VCA cues in on persons moving against the flow of traffic at security checkpoints as well as individuals repeatedly returning to the same point for no apparent reason. The real-time footage used for detection is also saved like conventional monitoring, and so can be used for later analysis.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01052234
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: BTRIS, TRIS
  • Created Date: Jun 24 2007 4:13PM