Performance Evaluation of Various Color Local Texture Models for Face Recognition

Texture is defined as the regular expression of an element or pattern on a surface. Due to its significance, this texture property is used in various applications incIuding Face Recognition. As Color local texture features are able to provide excellent face recognition rate when compared to grayscale texture features, the performance evaluation of various texture models Iike Color local binary pattern(CLBP), Color local Gabor wavelet(CLGW) and Local color vector binary pattern(LCVBP) for color face recognition are dealt in this paper. These three models exploits the discriminative information derived from spatiochromatic texture patterns of different spectral channels within a certain local region. Experiments were conducted under various challenges like different poses, expressions and various illuminations using the databases Iike IIT Kanpur, IMM and CMU-PIE. The results show that LCVBP gives better recognition rate than CLBP and CLGW.