Infrared face recognition based on LBP co-occurrence matrix and partial least squares

Infrared face recognition, being light-independent, and not vulnerable to facial skin, expressions and posture, can avoid or limit the drawbacks of face recognition in visible light. This paper proposes an infrared face recognition method based on Local Binary Pattern (LBP) co-occurrence matrix. In traditional LBP-based features such as LBP histogram, space locations information, which is an important feature for recognition, is discarded. Considering such spatial relations in infrared faces, we introduce co-occurrence matrix of LBP codes, instead of histogram, to obtain more discriminative representation for location features. To cope with the high dimensions of LBP co-occurrence matrix, the final classifier formulates Partial Least Squares (PLS) regression for accurate classification. The experimental results show combination of LBP co-occurrence matrix and PLS achieve better infrared face recognition performance compared to state-of-the-art approaches. Copyright © 2015 Inderscience Enterprises Ltd.