Face recognition based on fractional Fourier transform and PCA

Energy distribution over fractional Fourier transform (FRFT) can represent more characters of face images in different angles, which can improve recognition accuracy. This paper presents a novel method for face recognition based on FRFT and Principal component analysis (PCA). Firstly, the face images are transformed into FRFT domain. Then PCA is adopted to reduce the dimension of face images. In the end, Mahalanobis distance is used for classifying. Experimental results demonstrate that the proposed method outperforms traditional method. This provides new insights into the role that preprocessing methods play in dealing with images.