摘要
CAMSHIFT算法是一种基于颜色直方图的目标跟踪算法。在视频跟踪过程中,CAMSHIFT算法利用选定目标的颜色直方图模型得到每帧图像的颜色投影图,并根据上一帧跟踪的结果自适应调整搜索窗口的位置和大小,得到当前帧中目标的尺寸和质心位置。在介绍Intel公司的开源OpenCV计算机视觉库的基础上,采用CAMSHIFT跟踪算法,实现运动目标跟踪,解决了跟踪目标发生存在旋转或部分遮挡等复杂情况下的跟踪难题。实验结果表明该算法的有效性、优越性和可行性。
CAMSHIFT is an object CAMSHIFT operates on a color back tracking algorithm based on the color histogram. In the process of object tracking, projection image produced from object histogram model in current frame and finds the location and size of the current frame by adaptively adjusting the size and the location of the searching windows according to the tracking results of the previous frame in the video. On the basis of introducing OpenCV（an Intel open source computer vision library）, through CAMSHIFT algorithm, the paper realizes moving object tracking and resolves some problems including distractor and occlusion by other objects. Experimental results show good performances,superiority and feasibility of the algorithm.
出处
《现代电子技术》
2008年第20期128-130,138,共4页
Modern Electronics Technique
作者简介
李振伟 男，1983年出生，河南安阳人，硕士研究生。研究方向为视频运动目标跟踪。
陈翀 硕士研究生。
赵有 研究员、博士生导师。
