Face recognition based on adaptive singular value decomposition in the wavelet domain

Abstract

Face recognition is challenging because of lighting variation. This paper proposes a wavelet-based method combined with singular value decomposition (WSVD), which can be used to enhance face images, to overcome this problem. With the designated Gaussian template, the three color channels of the face image were next transformed to the discrete wavelet domain. By multiplying the singular value matrices of these frequency subband coefficient matrices with their corresponding compensation weight coefficients, the frequency subband coefficients of the three color channels were automatically adjusted. The 2D inverse discrete wavelet transform was then performed to obtain the WSVD-compensated color face image. The public color face databases confirmed that our proposed method can be efficiently applied in real applications.