关键词: 掌纹识别, Gabor小波变换, 主分量分析, 线性判别分析

Abstract: A feature extraction method for palmprint image is proposed，the implementation procedure of this method is as follows：firstly，the 2D Gabor wavelet transform coefficient amplitudes are computed at the equispaced discrete positions on the palmprint image，and they are used as original features of the palmprint image；then，the dimension of the Gabor wavelet feature is reduced by principal component analysis；lastly，the optimal discriminant features that are most advantageous for classification are extracted by linear discriminant analysis.Experimental results show the effectiveness of this method.

Key words: palmprint recognition, Gabor wavelet transform, principal component analysis, linear discriminant analysis