New approach to vehicle tracking based on region processing

Vehicle tracking is important in traffic monitoring systems. The behaviors of regions of moving vehicles are complicated, since the regions may combine or break during the tracking due to mistakes in vehicle detection and tracking or vehicles\´ overlapping with each other, and as a result, region matching simply according to similarities between successive frames is not enough to achieve reliable results. In this paper, a vehicle tracking method is proposed to reduce mistakes in spatial segmentation. "Temporary vehicle", "confirmed vehicle" and the corresponding judging rules are presented. In addition, fuzzy judgment is proposed to determine whether vehicle overlapping occurs or not. The situation of vehicle overlapping is classified into six types so as to adopt different rules for vehicle tracking. Experimental results show the efficiency of the proposed methods under complicated traffic scenes.