Complex kernel PCA for multimodal biometric recognition

This letter presents a novel multimodal biometric recognition algorithm based on complex kernel principle component analysis (CKPCA). CKPCA generalizes kernel principle component analysis (KPCA) method for complex field to perform feature fusion and classification. Iris and face are used as two distinct biometric modals to test our algorithm. Experimental results show that the proposed algorithm achieves much better performance than other conventional multimodal biometric algorithms.

[1]  Alex Pentland,et al.  Face recognition using eigenfaces , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[2]  Xiao-Yuan Jing,et al.  Face and palmprint feature level fusion for single sample biometrics recognition , 2007, Neurocomputing.

[3]  Jian Yang,et al.  Feature fusion: parallel strategy vs. serial strategy , 2003, Pattern Recognit..

[4]  Dexin Zhang,et al.  Personal Identification Based on Iris Texture Analysis , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[6]  Jian Yang,et al.  Combined Fisherfaces framework , 2003, Image Vis. Comput..

[7]  M. Tistarelli,et al.  Robust Feature-Level Multibiometric Classification , 2006, 2006 Biometrics Symposium: Special Session on Research at the Biometric Consortium Conference.