这是一个多重脸识别的程序，代码采用k_L变换和奇异值分解，利用协方差矩阵获得人脸图像的特征脸。-This is a multi-face recognition program, the code used k_L transform and singular value decomposition, the covariance matrix can be obtained by using the face image features the face.

估计类均值向量和协方差矩阵的方法。 类间离散度矩阵、类内离散度矩阵的计算方法 -Estimated class mean vector and covariance matrix method. Between-class scatter matrix, within-class dispersion matrix calculation method

协方差跟踪，将视频变成帧读入，计算协方差矩阵，并通过遍历找到最相似点区域进行跟踪，并生成轨迹-The covariance tracking, the video frame into read, calculate the covariance matrix, and find the similarities by traversing the area for tracking, trajectory generation and