A Simple Kernel Method with Shannon Wavelet Kernel for Face Recognition

To avoid using all training face images to recognize a test face image, the class centroids can be used; to alleviate the impairment of recognition accuracy due to face occlusion, Shannon wavelet kernel can be utilized. This paper presents a straight forward kernelbased approach with Shannon wavelet kernel for face recognition, which first finds the centroid face image in the kernel feature space for each class, and then compares test images with every kernel class centroid to decide which they belong to. Theoretical analysis and experimental results on Yale face database both show that our method is simple but capable of fast face recognition while maintaining high recognition accuracy.

[1]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[2]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[3]  Jian-Huang Lai,et al.  Face classification based on Shannon wavelet kernel and modified Fisher criterion , 2006, 7th International Conference on Automatic Face and Gesture Recognition (FGR06).

[4]  Hua Yu,et al.  A direct LDA algorithm for high-dimensional data - with application to face recognition , 2001, Pattern Recognit..

[5]  Rama Chellappa,et al.  Discriminant Analysis for Recognition of Human Face Images (Invited Paper) , 1997, AVBPA.

[6]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  B. Scholkopf,et al.  Fisher discriminant analysis with kernels , 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468).

[8]  Gunnar Rätsch,et al.  Invariant Feature Extraction and Classification in Kernel Spaces , 1999, NIPS.