Face Detection and Recognition Using Enhanced- LBP Feature Extraction in Digital Images

We present a new approach to face detection and recognition from digital images which considers both shape and texture information to represent face images. The face area in any digital image is first divided into small regions from which Enhanced Local Binary Pattern (E-LBP) histograms are extracted and concatenated into a single, spatially enhanced feature histogram efficiently representing the face image. The recognition is performed utilizing a closest neighbor classifier as a part of the figured element space with Chi square as a difference measure. Extensive experiments clearly show the superiority of the proposed scheme over all considered methods (PCA, Bayesian Intra/extra personal Classifier and Elastic Bunch Graph Matching) on FERET tests which include testing the robustness of the method against different facial expressions, lighting and aging of the subjects. In addition to its efficiency, the simplicity of the proposed method allows for very fast feature extraction.