Abstract:
For single-modal biometric system is susceptible to interference in appliacation, with low recognition rate, and not able to achieve zero error identification, a new fusion recognition approach in feature level of face and iris is proposed, based on the second-generation Curvelet and 2D Log-Gabor filtering.In the proposed approach, the second generation Curvelet is employed to extract face information, and amplitudes of 2D Log-Gabor are used to extract iris information.Then we use PCA to reduce the dimention of single-modal feature vectors, combine them in feature level, and distinguish fusion feature vectors by SVM.Experimental results on ORL face database and CASIA iris database show that: the correct fusion recognition rate can reach 100%, improved both 3.33% compared with single face feature and single iris feature, and the proposed algorithm is an effective model for multimodal biometric recognition.