中文说明：包含特征值与特征向量的提取、训练样本以及最后的识别，采用加权网络中节点强度和权重都是幂率分布的模型，IMC-PID是利用内模控制原理来对PID参数进行计算，最终的权值矩阵就是滤波器的系数，表示出两帧图像间各个像素点的相对情况，多元数据分析的主分量分析投影。
English Description：
Contain eigenvalues and eigenvector extraction, training samples and the final recognition, the weighted network node strength and weight are power-law distribution model, IMC-PID is by using the internal model control principle of PID parameters are calculated and final weight matrix is that of the filter coefficients, expressed between two adjacent frames of each pixel of the relative, multivariate data analysis principal component projection analysis.