J4 ›› 2009, Vol. 26 ›› Issue (6): 647-653.

• Image and Information Proc. • Previous Articles     Next Articles

Kernel Fisher discriminant analysis used in palmprint recognition

PEI Yu, LIU Hai-Lin   

  1. Faculty of Applied Mathematics, Guangdong University of Technology, Guangzhou 510006, China
  • Online:2009-11-03
  • Contact: 刘海林(1963-), 男, 河南安阳市人, 博士后, 教授, 研究方向为盲源分离, 进化计算。 E-mail:lhl@scnu.edu.cn

Abstract:

Kernel fisher discriminant analysis(KFDA) method is a more prominent method in pattern recogniztion to extract non-linear characteristics. We introduced kernel fisher discriminal analysis in the palmprint recognition to extract non-linear characteristics. First of all, we used wavelet transform to reduce palmprint images dimension based on retaining the original image information and features. Then, we used kernel Fisher discriminant analysis to extract features and the null-space KFDA method(ZKFDA) was introduced to solve the problem of small samples. Finally, we used a classifier to palmprint match based on minimum distance. Experimental results show that KFDA performs better than Two-Dimensional FLD(2DFLD) when the principal component numbers are different. ZKFDA performs better than KFDA in the average recognition rate, and computation has been significantly decreased. The recognition performance of radial basis function is the best in the selection of kernel functions.

Key words: image processing, Kernel Fisher discriminant analysis, feature extraction, palmprint recognition