遮挡情况下的视觉目标跟踪方法研究

将目标整体相关匹配算法和目标各子块相关匹配作表决的算法相结合,有效解决了运动目标被遮挡的跟踪问题.目标被遮挡,表现为某些子块被遮挡且匹配错误.对被遮挡的子块使其不参与表决,也不参与整体相关匹配的计算,只利用目标剩余的能代表目标本身属性的未遮挡子块继续跟踪目标.实验结果表明,采用的两种算法互为补充,对解决遮挡情况下目标的视觉跟踪是有效的.

Study on Vision Target Tracking Method under Occlusion

A method is proposed to solve target tracking problem of the moving object under occlusion efficiently in rigid target tracking by combining integral-correlation-matching algorithm with multi-block-voting algorithm. If the target is occluded, it means that some blocks are occluded and matching falsely. The occluded blocks are not to participate in voting and integral-correlation-matching calculation, and only the remainder unoccluded blocks th~.t can represent the target attribute are used to track the target unceasingly. Experimental results show that the adopted two algorithms are complementary, and effective combination of the two algorithms can achieve reliable vision tracking performance under the case of heavy occlusion of the moving objects.