Collaborative Kalman filters for vehicle tracking

摘要

Airborne vehicle tracking system is receiving increasing attention because of its high mobility and large surveillance scope. However, tracking multiple vehicles simultaneously on airborne platform is a challenging problem, owing to uncertain vehicle motion and visible frame-to-frame jitter caused by camera vibration. To address these problems, a new collaborative tracking framework is proposed. The framework consists of two level tracking processes: to track vehicles as groups, the higher level builds the relevance network and divides target vehicles into different groups; the relevance is calculated based on the status information of vehicles obtained by the lower level. This kind of group tracking takes into account the relevance of vehicles and reduces the impact of camera vibration, so the proposed method is applicable for multi-vehicle tracking in airborne videos. Experimental results demonstrate that the proposed method has better performance in terms of the tracking speed and accuracy compared to other existing approaches.