Object Tracking System for Traffic Surveillance

This paper proposes a technique for identifying a moving object in a video clip of stationary background for real time system and discusses one application that is traffic surveillance. We present a framework for detecting some important but unknown knowledge like vehicle identification and traffic flow count. The objective is to monitor activities at traffic intersections for detecting congestions, and then predict the traffic flow which assists in regulating traffic. Dynamic objects detection involves vehicle detection through background foreground segmentation and TLD. Tracking-Learning-Detection (TLD) is novel system for long-term tracking of objects in unconstrained videos and is built on the fly. The system extends TLD with the concept of a generic detector and a validator which is designed for real-time vehicle tracking. The off-line trained detector localizes the structure of vehicle given by vehicle detection and classification algorithm and the online trained validator decides which objects correspond to the tracked subject.