Contourlet-based feature extraction with PCA for face recognition

Face recognition is still a challenging task because face images can vary considerably in terms of facial expressions, lighting conditions,... etc. It is commonly known that the use of multiresolution filter banks improve the recognition accuracy of image based biometric systems. In this paper we propose to investigate the usefulness of the multiscale and directionality properties of the Contourlet Transform with a view to extract more discriminant features in order to further enhance the performance of the well known Principal Component Analysis method when applied to Face recognition. The proposed method has been extensively assessed using two different databases: the YALE face Database and the FERET Database. A series of experiments have been carried out and a comparative study suggests the efficiency of the Contourlet Transform in enhancing the classification rates of a number of known face recognition algorithms.