An Improved Face Recognition Algorithm Using Quantized DCT Coefficients

In this paper, we propose an improved face recognition algorithm using quantized Discrete Cosine Transform (DCT) coefficients for facial image recognition. For adding spatial information of the face to improve recognition performance, a region-division (RD) method is utilized. The facial area is firstly divided into 5 regions relating to the facial parts (forehead, eye, nose, mouth, jaw). Then feature vectors of each facial part are generated by using DCT coefficients in low frequency domains. Code vector referred count histogram, which is utilized as a very effective personal feature value, is obtained by Vector Quantization (VQ) processing. Recognition results with different parts are first obtained separately and then fused by weighted averaging. Experimental results show face recognition using proposed feature vector is very efficient.