Face recognition using MB-LBP and PCA: A comparative study

Abstract

Face recognition is an emergent research area, spanning over multiple disciplines such as image processing, computer vision and signal processing. Moreover, face recognition is also used for identity authentication, security access control and intelligent human-computer interaction. This work compares face recognition methods using local features and global features. The local features were derived using Multi Scale Block Local Binary Patterns (MB-LBP) and global features are derived using Principal Component Analysis (PCA). For each facial image a spatially enhanced, concatenated representation was obtained by deriving a histogram from each grid of the divided input image. These histograms were projected to lower dimensions by applying PCA which represents local features to characterize the face of a subject. The global face representation of a subject was derived by projecting several images of the subject into lower dimensions applying PCA. Face Recognition was performed with different similarity metrics on ORL, JAFFE and INDIAN face databases and compared with other works. It was found that the local features (MB-LBP) are better than the global features (PCA) for face recognition.