Video Tracker System for Traffic Monitoring and Analysis

In this study, a new algorithm is proposed to solve the multi-vehicle tracking problem that can deal with problems such as occlusion, short period object lost or inaccurate object detection. The scene, which is inspected in this study, is an uncontrolled junction with high traffic density where the above defined problems occur frequently. The developed system is a system that collects a-priory information about the junction, models that a-priory information and uses it in the tracking stage. The a-priory information collecting stage of the system is done automatically. Two different tracking methods are used together in the developed system: The multi-model Kalman tracker and the Markov scene partition tracker. The Markov scene partition tracker is used to increase the performance of the multi-model Kalman tracker. By this combination of the two trackers the continuity of the track is achieved for situations such as object loss and occlusion.