Camshift Target Tracking Algorithm Based on NCC Matching
ZHENG Zhaohui;Department of Public Courses,Wuhan Railway Vocational College of Technology;
Target tracking is always a hot topic of machine vision,usually target tracking determines the position of target by searching the similarity of upper and lower frames. Camshift algorithm commonly utilizes the color information of the target in the target tracking process,but the target tracking is easily failed when exists interference of similar color. While the NCC algorithm can use the structure features of target. Therefore,a tracking algorithm which combines Camshift with NCC is proposed,in which Camshift is used to locate the positioning of target,meantime,NCC matching is conducted by using the scaling in location area,then the final position of target is obtained. The experimental results show that the algorithm is feasible and effective,and the tracking performance is improved significantly compared with the traditional tracking algorithm.