一种结合背景差分的改进CamShift目标跟踪方法

针对传统的CamShift目标跟踪算法在跟踪目标之前需要手动选择目标区域,以及当目标物的颜色与背景色相似时容易跟丢的问题,提出了一种结合背景差分的CamShift改进方法.首先利用背景差分法检测出的目标区域作为CamShift算法的初始搜索窗,以解决手动选择跟踪目标的问题;然后在跟踪的过程中比较CamShift算法每一帧搜索窗的中心横纵坐标与背景差分法检测出的目标区域矩形框中心的横纵坐标,根据设定的阈值来选择CamShift算法下一帧的搜索窗,以解决目标色与背景色相似的问题.仿真实验表明改进方法在目标与背景色相似的情况下也能有效地跟踪目标.

On a Combination of Background Difference Improved CamShift Target Tracking Method

In view of the traditional CamShift target tracking algorithm in tracking the target before needing to manually selected the target area,when the color of the target and the background color similar goals easily lost,a combination of background difference method to improve the CamShift tracking method has been proposed.First-ly,the target area as the initial search window of CamShift algorithm by background difference method detected to solve the problem of manually choose target tracking.Then,in the process of tracking,a comparison of CamShift algorithm search window at the center of the horizontal ordinate each frame and background difference method has been used to detect the target area of the rectangular box center horizontal ordinate,according to the setting threshold to choose CamShift algorithm the search window of next frame,in order to solve problems that the color similarity of target and background.Finally,experimental results show that the method in the target color and background color similar cases can well track the target object.