首页 | 本学科首页   官方微博 | 高级检索  
   检索      

一种改进的Mean-shift目标跟踪算法
引用本文:要俊杰,ZHANG Gui-min,宋爱斌,LIU Ying.一种改进的Mean-shift目标跟踪算法[J].装甲兵工程学院学报,2008,22(3).
作者姓名:要俊杰  ZHANG Gui-min  宋爱斌  LIU Ying
作者单位:装甲兵工程学院,基础部,北京,100072
摘    要:提出了一种利用多帧图像中目标区域颜色直方图特征信息的Mean-shift目标跟踪算法,该算法以Mean-shift为核心,在对目标模板建立特征模型时,不仅考虑了前一帧的目标区域的颜色直方图,而且将前数帧中目标区域的颜色直方图通过加权一起考虑,从而提高了跟踪的稳定性。

关 键 词:直方图  目标跟踪  Mean-shift

An Improved Mean-shift Algorithm for Object Tracking
YAO Jun-jie,ZHANG Gui-min,SONG Ai-bin,LIU Ying.An Improved Mean-shift Algorithm for Object Tracking[J].Journal of Armored Force Engineering Institute,2008,22(3).
Authors:YAO Jun-jie  ZHANG Gui-min  SONG Ai-bin  LIU Ying
Institution:1(1. Department of Fundamental Courses; Academy of Armored Force Engineering; Beijing 100072; China; 2. Department of Control Engineering; China);
Abstract:Based on the color histogram information of target region in multi-frame images, a new Mean-shift algorithm for object tracking is put forward. Focusing on Mean-shift, when establishing characteristic model for target template, the new algorithm not only considers the color histogram of target region in the proceeding frame, but also considers the color histogram of target region in the first several frames through the weighting. Therefore, the stability of tracking is improved.
Keywords:histogram  object tracking  Mean-shift
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号