摘要
Camshift(Continuously Adaptive Mean Shift)算法具有很好的实时性和鲁棒性,但是当目标遇到大面积的类目标颜色干扰,目标被严重遮挡时跟踪会失败。针对这些问题,提出ABCshift(Adaptive Background Camshift)结合Kalman滤波的改进算法。实验表明所提出的算法能有效解决以上问题,在复杂的背景下有良好的适应性,并与其他改进的Camshift算法进行了对比。
Camshift algorithm has good real-time property and robustness,but its tracking may fail when the object is interfered by the colour similar to it and in large area or the object is occluded seriously.To solve these problems,we present an improved algorithm which combines the ABCshift with Kalman filtering.Experimental results show that the proposed algorithm can overcome the above problems effectively and has good adaptability in complex background;moreover,this one is compared with other improved Camshift algorithms.
出处
《计算机应用与软件》
CSCD
北大核心
2014年第2期226-229,共4页
Computer Applications and Software
基金
国家自然科学基金项目(60972158)
中南民族大学中央高校基本科研业务费专项基金项目(czy12012)