利用openCV，首先利用goodFeaturesToTrack函数得到图像中的强边界作为跟踪的特征点，接下来要调用calcOpticalFlowPyrLK函数，输入两幅连续的图像，并在第一幅图像里选择一组特征点，输出为这组点在下一幅图像中的位置。再把得到的跟踪结果过滤一下，去掉不好的特征点。再把特征点的跟踪路径标示出来。(Using openCV, we use goodFeaturesToTrack function to get strong edges in the image as the feature point tracking, next to call calcOpticalFlowPyrLK function, input two consecutive images, and select a set of feature points in the first image, the output for this set of points in the next image position. Then filter the resulting trace to get rid of the bad feature points. Then mark the tracking path of the feature points.)