基于核局部Fisher判别分析的掌纹识别

Palmprint recognition based on kernel localized Fisher discriminant analysis

运用核局部Fisher判别分析(KLFDA)进行掌纹识别。为了解决小样本图像识别中特征方程矩阵的奇异性问题,首先运用图像下抽样方法降低掌纹空间的维数,在低维图像上应用KLF-DA提取低维的投影向量;然后将训练图像和待识别图像的核矩阵向投影向量上投影,得到非线性局部判别特征;最后计算特征向量间的余弦距离,进行掌纹匹配。运用PolyU掌纹图像库对算法进行测试,实验结果表明,与主元分析(PCA)、Fisher判别分析(FDA)、独立元分析(ICA)、核主元分析(KPCA)和局部Fisher判别分析(LFDA)相比,本文算法的识别率(RR)最高为99%,特征提取和匹配总时间0.031 s,满足实时系统的要求。

A new palmprint recognition method based on kernel localized Fisher discriminant analysis （KLFDA） is proposed. In order to solve the singularity of the eigenvalue equation matrix in small-sizesample cases, such as image recognition, the image down-sample is first used to reduce palmprint space dimensionality. The KLFDA is applied to extract the low projection vectors. Then the kernel matrices of the training images and test images are projected onto the projection vectors to get the nonlinear localized palmprint feature vectors. Finally,the cosine distance between two feature vectors is calculated to match palmprint. The new algorithm is tested in PolyU plmprint database. The results show that compared with principal component analysis （PCA）, Fisher discriminant analysis（FDA）,independent component analysis （ICA）,kernel principal component analysis （KPCA） and localized Fisher discriminant analysis （LFDA） ,the recognition rate （RR） of the new algorithm is the highest,which is 99% ,and the total time for feature extraction and matching is 0. 031 s, so it meets the real-time system specification.