A Contourlet-domain Feature Extraction Scheme for Face Recognition

In this paper, a new multi-resolution tool, namely contourlet transform, is used to extract features for face recognition, which efficiently exploits the local spatial variations in a face image. Multi-resolution ideas, notably the wavelet transform, have been profusely employed for addressing the problem of face recognition. However, theoretical studies indicate that digital contourlet transform is an even better method than wavelets due to its directional properties. For the purpose of feature extraction, instead of considering the entire face image, an entropy-based local band selection criterion is developed, which selects highinformative horizontal segments from the face image. In order to capture the local spatial variations within these highinformative horizontal bands precisely, the horizontal band is segmented into several small spatial modules. Dominant contourlet transform coefficients corresponding to each local region residing inside those horizontal bands are selected as features. In the selection of the dominant coefficients, a threshold criterion is proposed, which not only drastically reduces the feature dimension but also provides high withinclass compactness and high between-class separability. Extensive experimentation is carried out upon standard face databases and a very high degree of recognition accuracy is achieved by the proposed method in comparison to those obtained by some of the existing methods.