中文说明：LDA线性判别分析是一种经典的提取特征的算法，它的基本思想是通过样本的类内离散度和类间离散度，寻找由最佳投影矢量构成的投影矩阵。将原始的样本数据投影到特征子空间中，实现数据分类。由于在人脸识别时常常会遇到小样本问题，因此在本次代码中，先用PCA主成分分析的方法降低样本维数，再用线性判别分析提取特征。最后，使用最小距离分类器判别分类，得到一个投影维度和识别率的曲线图。
English Description：
LDA linear discriminant analysis is a classical feature extraction algorithm. Its basic idea is to find the projection matrix composed of the best projection vector through the within class and inter class dispersion of samples. The original sample data is projected into the feature subspace to realize data classification. Because we often encounter the problem of small samples in face recognition, in this code, PCA principal component analysis is used to reduce the dimension of samples, and then linear discriminant analysis is used to extract features. Finally, the minimum distance classifier is used to judge the classification, and a graph of projection dimension and recognition rate is obtained.