Background subtraction algorithms for a video based system

[摘要] ENGLISH ABSTRACT: To reliably classify parts of an image sequence as foreground or backgroundis an important part of many computer vision systems, such as video surveillance,tracking and robotics. It can also be important in applications wherebandwidth is the limiting factor, such as video conferencing.Independent foreground motion is an attractive source of information for thistask, and with static cameras, background subtraction is a particularly populartype of approach. The idea behind background subtraction is to comparethe current image with a reference image of the background, and from theredecide on a pixel by pixel basis, what is foreground and what is backgroundby observing the changes in the pixel sequence.The problem is to get the useful reference image, especially when large partsof the background are occluded by moving/stationary foreground objects; i.e.some parts of the background are never seen.In this thesis four algorithms are reviewed that segment an image sequenceinto foreground and background components with varying degrees of successthat can be measured on speed, comparative accuracy and/or memory requirements.These measures can be then effectively used to decide the applicationscope of the individual algorithms.

[发布机构] Stellenbosch University