本程序是在vs2010+opencv的平台上运行的，利用的是opencv2.4.9，C++写的，该算法与传统跟踪算法的显著区别在于将传统的跟踪算法和传统的检测算法相结合来解决被跟踪目标在被跟踪过程中发生的形变、部分遮挡等问题。同时，通过一种改进的在线学习机制不断更新跟踪模块的“显著特征点”和检测模块的目标模型及相关参数，从而使得跟踪效果更加稳定、鲁棒、可靠。最后的得到的效果令人满意。-This program is run on a platform vs2010+opencv, the use of the opencv2.4.9, C++ written significant difference between this algorithm and the traditional tracking algorithm is the traditional tracking algorithm combining traditional detection algorithm to solve tracked targets It is deformed during tracking, partial occlusion and other issues. At the same time, through an improved online learning mechanism has been updated tracking module object model and related parameters " significant feature points"　and the detection module, so that tracking was more stable, robust and reliable. The final was satisfactory.