Abstract: Aiming at the problem that the current target tracking algorithm was subject to large environmental interference and poor anti\|occlusion ability, the author
designed an anti\|occlusion tracking algorithm for moving objects. Firstly, the limitations of the classical tracking algorithm for moving objects—mean
 shift algorithm were analyzed, and then salient features that contributed most to the tracking were selected from the target candidate features, and the adaptive template updating strategy was adopted to improve the accuracy of the moving objects. The simulation results show that the algorithm can track moving object accurately and realtime, accelerate the tracking speed of moving object,  have good robustness to occlusion object, and obtain better tracking results for moving object.

Key words: robustness test, occlusion ability, anti\, mean shift algorithm, moving object tracking, template updating