Research on Moving Object Tracking under Occlusion Conditions

With the development of information fusion, a new theory based on Dezert-Smarandache theory (DSmT) was widely applied to lots of aspects. Owing to the difficulty of handling conflict evidence combination when tracking multiple moving objects under occlusion conditions, a DSmT method with anti-occluding objects tracking in complex background is proposed based on occlusion judgment prediction, and the corresponding computing examples of generating hyper-power sets and tracking under different occluding conditions are analyzed and discussed in this paper. Calculation results show that the method is proven to be correct and effective in tracking occluded objects. When analyzing multi-objects tracking under very complex environment, it will lay a firm foundation for further solving information fusion of conflict evidences.