Vision-Based Detection and Tracking of Moving Target in Video Surveillance

In this paper a real-time detection and tracking of moving targets is presented. The scheme involved four phases. Phase one: Object segmentation which used to identify the foreground objects from the background by using background subtraction based on temporal differencing and finding the average background model. Phase two: Object recognition used to identify the foreground objects that should be tracked by using simple blob detection. Phase three: Object representation which takes the outcome from phase two. It computes the recognized object to be tracked. Phase 4: Object tracking that used Kalman filter. The results show that the tracking system is capable of target shape recovery and therefore it can successfully track targets with varying distance from camera or while the camera is zooming.