Face recognition based on a novel illumination normalization method

A novel illumination normalization method is proposed in this paper to deal with illumination variations in face recognition. Firstly, Discrete cosine transform (DCT) is used on the original images in logarithm domain. Direct current (DC) coefficient is reset and some low frequency alternating current (AC) coefficients are set as zero to eliminate illumination variations in large areas. Secondly, local normalization method is used on the inverse discrete cosine transform images to minimize illumination variations in small areas. Experimental results show that the proposed method can eliminate effect of illumination variations effectively and improve performance of face recognition methods significantly. The proposed method be used to eliminate the effect of illumination variations before face recognition.