基于区域形心及其矢量方向的视频对象跟踪

提出了一种新的基于区域形心及其矢量方向的视频对象跟踪算法。视频对象分割后，使用区域增长法将视频对象分成若干个区域，然后利用视频序列连续两帧中每个区域的形心及形心间矢量的方向，快速、有效地自动跟踪多个目标对象。实验结果表明，该算法对运动物体的出现和消失，以及非刚性物体的尺度变化和变形，具有较强的鲁棒性。

Tracking video object based on regional-centroid and vector-direction

An algorithm was proposed to track moving objects based on centroid and vector in this paper. After segmenting video objects, video objects were divided into several regions by region-growing method. Then with the help of centroid and direction of vector of each region in two successive frames of a video sequence, it is able to track multi-nonrigid moving objects fast, effectively and automatically. The experimental results show that the presented method is robust to deal with the problems in multi-targets tracking such as new target entering, target leaving, scaling and deforming of nonrigid objects.