Face Recognition by Nonsubsampled Contourlet Transform and Support Vector Machine

摘要

This paper proposed a new face recognition method based on nonsubsampled contourlet transform (NSCT) and component-based support vector machine (SVM). This approach takes technological advantages of both SVM for classification and the NSCT for facial feature extraction. We named the new method combining the nonsampled contourlet transform and SVM as NSCT-SVM. The experimental results on ORL and MIT-CBCL face database demonstrate that our face recognition method is effective and competitive.