Illumination Invariant Face Recognition using SQI and Weighted LBP Histogram

Face recognition under uneven illumination is still an open problem. One of the main challenges in real-world face recognition systems is illumination variation. In this paper, a novel illumination invariant face recognition approach base on Self Quotient Image (SQI) and weighted Local Binary Pattern (WLBP) histogram has been proposed. In this system, the performance of the system is increased by using different sigma values of SQI for training and testing. Furthermore, using two multi-region uniform LBP operators for feature extraction simultaneously, made the system more robust to illumination variation. This approach gathers information of the image in different local and global levels. The weighted Chi square statistic is used for histogram comparison and NN (1-NN) is used as classifier. The weighted approach emphasizes on the more important regions in the faces. The proposed approach is compared with some new and traditional methods like QI, SQI, QIR, MQI, DMQI, DSFQI, PCA and LDA on Yale face database B and CMU-PIE database. The experimental results show that the proposed method outperforms other tested methods.