Face recognition using wavelet transform and Kernel Principal Component Analysis

A novel face recognition method using wavelet transform and Kernel Principal Component Analysis (KPCA) was presented. The method calculated logarithm transform and 2-dimensional wavelet transform for face pre-processing, used KPCA algorithm for face feature extraction, and adopted nearest neighborhood classifier based on Cosine distance for feature classification. The experimental results on Yale B frontal face database show that the face recognition rate of the proposed method can attain 100%. That is, the proposed approach can alleviate variable illumination for face recognition and identify all test samples on Yale B frontal face database accurately..