Python implementation of PCA face recognition

These days to see the PCA and face recognition of the process, and search the corresponding Python code on the network, there is, but the code quality is not good, so they re-write the next, for the att_faces data set recognition rate can reach 92.5%~98.0% (40 types, each randomly selected 5 training, 5 recognition), all code below, less than 50 lines oh.

#-*-Coding:utf-8-*-import numpy as Npimport os, glob, Random, Cv2def PCA (data,k): data = Np.float32 (Np.mat (data))    Rows,cols = data.shape #取大小 Data_mean = Np.mean (data,0) #求均值                                       Z = Data-np.tile (Data_mean, (rows,1)) D,v = Np.linalg.eig (z*z.t) #特征值与特征向量 V1 = v[:,: K] #取前k个特征向量 V1 = Z.t*v1 for i in Xrange (k): #特征向 Volume normalization V1[:,i]/= np.linalg.norm (V1[:,i]) return Np.array (Z*V1), Data_mean,v1def loadimageset (folder=u ' e:/Thunderbolt download/face Process/att_faces ', samplecount=5): #加载图像集, randomly select Samplecount images for training traindata = []; TestData = []; Ytrain=[];    Ytest = []; For k in range: Folder2 = os.path.join (folder, ' s%d '% (k+1)) data = [Cv2.imread (D.encode (' GBK '), 0) for D in Glob.glob (Os.path.join (Folder2, ' *.PGM '))] Sample = Random.sample (range (ten), Samplecount) Traindata.ext End ([Data[i].ravel () For I in range (ten) if I in Sample]) Testdata.extend ([Data[i].ravel ()-I in range () If I is not in sample]) Ytest.extend ([k]* (10-samplecount)) ytrain.extend ([k]* Samplecount) return Np.array (Traindata), Np.array (YTra IN), Np.array (TestData), Np.array (Ytest) def main (): Xtrain_, Ytrain, xtest_, ytest = Loadimageset () Num_train, num_ Test = Xtrain_.shape[0], xtest_.shape[0] xtrain,data_mean,v = PCA (Xtrain_,) xTest = Np.array ((Xtest_-np.tile ( Data_mean, (num_test,1))) * V) #得到测试脸在特征向量下的数据 ypredict =[ytrain[np.sum ((Xtrain-np.tile (d, (num_train,1)) **2, 1). Argmin ()] for D in xTest] print U ' European distance method recognition rate:%.2f%% '% ((ypredict = = Np.array (ytest)). Mean () *100) SVM = Cv2. SVM () #支持向量机方法 Svm.train (Np.float32 (Xtrain), Np.float32 (ytrain), params = {' Kernel_type ': C V2. Svm_linear}) ypredict = [Svm.predict (d) for D in Np.float32 (xTest)] #yPredict = Svm.predict_all (Xtest.astype (Np.floa T64)) print U ' support vector machine recognition rate:%.2f%% '% (Ypredict = = Np.array (ytest)). Mean () *100) if __name__ = = ' __main__ ': Main () 

　　

Python implementation of PCA face recognition