The Research Of Tracking For Fast Moving Objects Based On Mean Shift

Target tracking is very active in the field of computer vision a subject,has a very wide range of applications,such as video conference call,robot navigation,and simulate the reality and so on.The performance of the moving target tracking algorithm directly affects the target tracking accuracy and robustness.At present in the existing target tracking algorithm,the tracking algorithm based on target characteristics or motion information,although able to achieve the purpose of target tracking,but need to deal with a large amount of data,operation is relatively complex,it is difficult to realize real-time tracking,even more in the fast moving target tracking,such as football.On the current research situation,this article before people,on the basis of the research work on no density estimation theory is studied,and the target tracking algorithm based on Mean Shift algorithm,in-depth studies are expected this algorithm in target tracking problem often encountered in real time and keep out in improvement.In this paper,all kinds of target tracking algorithm based on Mean Shift algorithm has summarized and their advantages and disadvantages and applicable scenario.No arguments in statistical density estimation theory were studied,the existing density estimation method of parameters and applicable conditions are summarized,based on this,also has no parameters based on density estimation theory of a statistical iterative method of Mean Shift algorithm has carried on the deep discussion.This paper also studied the convergence of Mean Shift algorithm,and the rigorous proof.In this paper,based on the Mean Shift tracking a moving target tracing algorithm is discussed in detail,and the traditional target tracking algorithm on the basis of several kinds of movement target tracking scheme were discussed:the Mean Shift tracking algorithm is combined with Kalman motion prediction for effective tracking scheme;Mean Shift tracking algorithm is combined with a centroid tracking algorithm.The simulation Kalman motion prediction and centroid algorithm was combined with a Mean Shift tracking algorithm for moving target tracking effectively.Tracking results show that this improved Mean Shift tracking algorithm of target tracking algorithm for fast moving object limitations and shade problem,has a good robustness.