眨眼检测，先用haar特征框定人脸，然后用camshift跟踪人脸，根据几何特征得到人眼的大概位置，然后根据直方图的变化检测眨眼-Blink detection, the first feature with the haar framed face, and then use camshift tracking human face, according to the geometric features are the approximate location of the human ey

camshift跟踪代码，基于opencv-camshift tracking code, based on opencv

采用 CAMSHIFT 算法快速跟踪和检测运动目标的 C/C++ 源代码，OPENCV BETA 4.0 版本在其 SAMPLE 中给出了这个例子-CAMSHIFT algorithm using fast-track and detect moving targets C/C++ source code, OPENCV BETA 4.0 version of the SAMPLE in this example is given

一个Camshitft算法，虽然opencv有Camshift函数，但是用matlab写能更好的了解原理。-A Camshitft algorithm, although there Camshift opencv functions, but using matlab to write a better understanding of the principles.

通过手动框选目标区域，然后进行自动跟踪，使用的算法是Camshift，自适应均值漂移(By manually selecting the target area, then automatic tracking, using the algorithm is Camshift, adaptive mean shift)