Combining Local Binary Pattern and Local Phase Quantization for Face Recognition

Combining information contained in image Spatial and frequency domains can provide rich important clues not seen in either individual of these domains. This paper presents a method of face recognition based on local binary pattern(LBP) and local phase quantization(LPQ). The face image is divided into several regions, LBP operator is used to extract LBP feature in the spatial domain and LPQ operator is used to extract LPQ feature in the frequency domain. Then concatenated into an enhanced feature vector to be used as a face descriptor. The simulation experiments illustrate that this method has better recognition rate and more robust than single method on the YALE and AR face database.