Combined with diagonal image transform, two-dimensional discrete cosine transform (2DDCT) is used in face and iris image for feature compression; then Kernel Fisher Discriminant Analysis (KFDA) is chosen as feature fusion; finally, Nearest Neighbor (NN) classifier is selected to perform recognition. Experimental results on ORL (Olivetti Research Laboratory) face database and CASIA (Chinese Academy of Sciences, Institute of Automation) iris database show that the dimension is reduced, the classified information is utilized, and correct recognition rate is improved effectively. A new approach is supplied for multimodal biometric identification.
