Construction and Application of SVM Model and Wavelet-PCA for Face Recognition

This work presents a method to increase the face recognition accuracy using a combination of Wavelet, PCA, and SVM. Pre-processing, feature extraction and classification rules are three crucial issues for face recognition. This paper presents a hybrid approach to employ these issues. For preprocessing and feature extraction steps, we apply a combination of wavelet transform and PCA. During the classification stage, SVMs incorporated with a binary tree recognition strategy are applied to tackle the multi-class face recognition problem to achieve a robust decision in presence of wide facial variations. The binary trees extend naturally, the pairwise discrimination capability of the SVMs to the multiclass scenario. Two face databases are used to evaluate the proposed method. The computational load of the proposed method is greatly reduced as comparing with the original PCA based method on the ORL and Compound face databases. Moreover, the accuracy of the proposed method is improved.