An efficient illumination compensation based on plane-fit for face recognition

Illumination variation is one of the critical factors affecting face recognition rate. A novel illumination compensation for face recognition is presented in this paper. First of all, a new illumination direction map is introduced, which is built using the block-based histogram equalization (Xudong Xie and Kin-Man Lam, 2005) and illumination model. Secondly, plane-fit method can be used to determine the category of the lighting source based on the illumination direction map, is first proposed. A corresponding model of illumination compensation is applied to convert uneven illumination face image to even illumination face image. The proposed algorithm has been evaluated based on the Yale database, the AR database, the YaleB database and the combined database. Experimental results show that, the method proposed can greatly improve the recognition rate and speed.