Multiscale illumination normalization for face recognition using dual-tree complex wavelet transform in logarithm domain

A new illumination normalization approach based on multiscale dual-tree complex wavelet transform (DT-CWT) is presented for variable lighting face recognition. In our method, low resolution wavelet coefficients are truncated to minimize variations under different lighting conditions. On the other hand, the coefficients in the directional subbands are used for extracting the invariant facial features. In order to reduce the halo artifacts, new conduction function is presented which can smooth the high-frequency illumination discontinuities while keeping most of the facial features unimpaired. The histogram remapping technique is employed on the normalized image which remaps the histogram into normal distribution and thus it can improve the recognition performance. Experimental results on the Yale B and CMU PIE databases show that the proposed method can achieve satisfactory performance for recognizing face images under large illumination variations.