PERFORMANCE METRICS IN VIDEO SURVEILLANCE SYSTEM

Abstract

Video surveillance is an active research topic in computer vision. One of the areas that are being actively researched is on the abilities of surveillance systems to track multiple objects over time in occluded scenes and to keep a consistent identity for each target object. These abilities enable a surveillance system to provide crucial information about moving objects behaviour and interaction. This survey reviews the recent developments in moving object detection and also different techniques and approaches in multiple objects tracking that have been developed by researchers. The algorithms and filters that can be incorporated in tracking multiples object to solve the occluded and natural busy scenes in surveillance systems are also reviewed in this paper. This survey is meant to provide researchers in the field with a summary of progress achieved up to date in multiple moving objects tracking. Despite recent progress in computer vision and other related areas, there are still major technical challenges that need to be solved before reliable automated video surveillance system can be realized.

Keywords

• Surveillance systems
• Object detection
• Object tracking
• Partial occlusion
• Crowded scenes