An Improved PCA Face Recognition Algorithm Based on the Discrete Wavelet Transform and the Support Vector Machines

In this paper, an improved PCA face recognition algorithm based on the Discrete Wavelet Transform and the Support Vector Machines is presented. The 2-D Discrete Wavelet Transform has been used to process the ORL standard face images to form the low frequency sub images by extracting the low frequency component. Then the PCA method is used to obtain the characterizations of these sub images. At last, the extracted eigenvectors are put into the SVM classifier for training and recognition. The experimental results indicate that this algorithm reduces the computational quantity because the dimension of the total population scatter matrix of the source images has deduced a lot and the performance of the SVM classifier is superior to many other classifiers. Compared with the traditional PCA face recognition algorithm, the calculation speed and the recognition efficiency here increase a lot.