Face Recognition against Varying Lighting Conditions Using Oriented Phase Congruency Image Features

Although face recognition has become a topic of active research in recent decades, an accurate face recognition is still a tough job and still struggle with its performance especially if it is unconstrained lighting variations. The external conditions present during the facial image acquisition stage deeply influence the appearance of a face in the acquired image and hence affect the performance of the recognition system. This paper presents an efficient method for robust face recognition against varying lighting condition by using Oriented Phase Congruency features. The extracted features, derived from the Gabor Phase Congruency response are concatenated to construct a feature vector to be used for classification by Subspace LDA. We demonstrate the effectiveness and superiority of our proposed method by conducting some experiments on Yale B face databases and our proposed method shows robust face recognition performance in the presence of severe lighting changes.