一种自适应模板更新的CamShift跟踪算法  

A CamShift Tracking Algorithm with Adaptive Model Update

作　　者：熊昊 段锦[1] 陈小远 代玉强 于津强 XIONG Hao, DUAN Jin, CHEN Xiao-yuan, DAI Yu-qiang, YU Jin-qiang(College of Electronic and Information Engineering, Changchun University of Science and Technology, Changchun 13000)

机构地区：[1]长春理工大学电子信息工程学院,长春130000

出　　处：《现代计算机》2018年第12期45-49,共5页Modern Computer

摘　　要：针对Cam Shift跟踪算法中跟踪窗口漂移和发散问题,提出加入SURF特征的改进算法。利用SURF特征尺度信息的变化约束搜索框,改善跟踪窗口在遭遇大面积背景色干扰时易发散的问题;利用SURF特征点数量、尺度和方向信息的变化选择相应的模板更新策略,使模板能更好地反映目标的特征,从而在下一帧中更准确地跟踪目标。Proposes an improved algorithm based on SURF feature for the problem of search window shifting and expanding in CamShift tracking algorithm.For the problem of the tracking windoweasily diverged in the similar color background,constraining search window by using of the SURF feature scale information.Chooses the corresponding template update strategy according to the change of the number,size and direction information of SURF feature points,so the templates will better reflect the characteristics of the target,and the tracking will be more accurate in the next frame.

关 键 词：CAMSHIFT算法 SURF 尺度 模板更新 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]