Face Recognition using Multiresolution PCA

In this paper we develop two techniques for face recognition using the idea of multiresolution face recognition. The multiresolution subbands are generated by using wavelet transform. The first technique is called the Multiresolution Feature Concatenation (MFC), where we use principal component analysis (PCA) as a dimensional reduction approach on each subband then concatenate the resulting projection coefficients of each subband together and perform classification. The second technique is called the Multiresolution Majority Voting (MMV), where the PCA approach and the classification are done separately on each subband and then the majority voting is applied for making decision. Both techniques show promising results and MMV approach outperforms the MFC approach. Moreover, the two techniques outperform the conventional PCA approach.