A novel infrared face recognition based on local binary pattern

To extract the discrimination local structural features, an improved infrared face recognition method based on LBP is proposed in this paper. To get robust local features in infrared face, local binary pattern representation is applied to our method, instead of holistic feature extraction method. Based on the criterion of separability discriminant, pattern selection (PS) algorithm is proposed to get the LBP patterns which are most suitable for infrared face recognition. To make full use of the space locations information, the partitioning and LBP histogram are applied to get final features. The experimental results demonstrate the infrared face recognition method based on LBP+PS proposed outperforms the traditional methods based on LBP or PCA+LDA.