Gary R．Bradski提出CamShift算法，即"Continuously Adaptive Mean-Shift"算法。是以颜色直方图为目标模式的目标跟踪算法，可以有效地解决目标变形和部分遮挡的问题，而且运算效率很高。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。-Gary R. Bradski CamShift proposed algorithm, that is, " Continuously Adaptive Mean-Shift"　algorithm. Color histogram for the target model is based on target tracking algorithms that can effectively solve the target deformation and partial occlusion of the problem, and efficient operation. In the video tracking process, CAMSHIFT algorithm uses color histogram model of the selected target by the color of each fr a me image projection, and according to the results on an adaptive tracking the location and size of the search window to get the target in the current fr a me size and centroid location.