基于混合高斯模型的Mean-Shift模板更新算法

针对基于Mean-Shift目标跟踪算法中遇到的不能对模板进行实时更新的问题,提出一个基于混合高斯背景建模的目标模板更新算法.该算法将目标视为背景,对目标中的每一个像素点利用三个高斯函数对它进行建模,利用每次Mean-Shift跟踪到的目标区域来对先前建立的混合高斯模型进行实时更新,将混合高斯模型得到的目标模板作为下一帧跟踪的目标模板.该算法较好地解决了基于Mean-Shift算法的模板更新问题,实验证明该算法是有效的.

AN ALGORITHM OF MEAN-SHIFT TEMPLATE UPDATE BASED ON MIXTURE GAUSSIAN MODEL

To improve the limitation in Mean-Shift object tracking algorithm which lacks real-time template update,an algorithm of target template updating based on mixture Gaussian background modelling is proposed.It treats the target region as "background",uses three Gaussian functions to model every pixel point in the target.The target regions tracked by the Mean-Shift algorithm each time are used to update in real-time the mixture Gaussian models set up previously,and the target template obtained from mixture Gaus...