Improved Camshift tracking method based on the edge suppression

Abstract

In order to improve the tracking performance of the target in complex background, a novel Camshift tracking method based on the edge suppression is proposed. Because the edge of the target is probably most vulnerable to the background, the gray value of the target edge in the back projection image is suppressed by the weighted method according to the position and size in the previous frame. Thus, the disturbance information from the background can be prevented from involving the target tracking. The method is applied to track the actual target and the target in the video of the standard test libraries. Experimental results show that the accuracy and stability of the target recognition and tracking can be improved by suppressing the information of the edge of the tracking target in the back projection image. And the method can also meet the real time tracking requirement of the TV tracking system.