CONSISTENT VEHICLES TRACKING BY USING A COOPERATIVE DISTRIBUTED VIDEO SURVEILLANCESYSTEM

ABSTRACT:This paper proposes a method of consistent vehicles tracking using a distributed cooperative multi-camera surveillance system.Vehicles’ matching in views of different cameras is a fundamental issue for increasing the accuracy of consistent tracking. Among the most important research in intelligent transportation systems (ITS), automatically urban traffic surveillance is one of the critical and challenging tasks. A single-camera system has inherent limitations such as limited field of view and errors in background subtraction. The proposed work addresses synchronizing the cameras for tracking vehicles simultaneously in overlapping fields of view. Multiple cameras with overlapping fields of view can provide fault tolerance and robustness for issues such as vehicles occlusion. Distributed fundamental purpose is to efficiently reduce the transmission rate and also analyze a traffic scene and report statics and information of interests. In this manner, the surveillance area is wider and using multiple views of cameras to produce uniform tracking can enhance the capability and performance of consistent tracking. By presenting anew transmission protocol in a robust manner, we evaluate the tracking accuracy. By this way, in occlusion situations, the vehiclesin overlapping fields of view are tracked more accurately.