核不相关空间算法及其在人脸识别中的应用

林玉娥; 顾国昌; 刘海波; 沈晶

doi:10.3724/SP.J.1146.2008.00718

核不相关空间算法及其在人脸识别中的应用

doi: 10.3724/SP.J.1146.2008.00718

基金项目:

中国博士后基金(20060400809)和黑龙江省青年科技基金(QC06C022)资助课题

计量
- 文章访问数: 3311
- HTML全文浏览量: 94
- PDF下载量: 742
- 被引次数: 0
出版历程
- 收稿日期: 2008-06-02
- 修回日期: 2009-03-30
- 刊出日期: 2009-08-19

Kernel Uncorrelated Space Algorithm and Its Application to Face Recognition

摘要

摘要: 不相关空间算法是求解不相关鉴别矢量集的快速算法，但是将其应用在人脸识别中将遇到小样本问题，并且算法只是一种线性的特征提取方法。该文提出一种核不相关空间算法，该方法的关键是高维特征空间中不相关空间的计算，对此提出一种简单的计算方法，即根据eigenface中将高阶矩阵计算转化成低阶矩阵计算的思想，将高维特征空间中不相关空间的计算仍归结为标准的特征值分解问题。所提出的算法能够有效地解决小样本问题。在ORL人脸库上的实验结果验证了所提出的算法的可行性和有效性。
- 人脸识别;不相关空间算法;小样本问题;核不相关空间算法
Abstract: Uncorrelated space algorithm is a fast method for the uncorrelated discriminant vectors extraction，but it may encounter the small size samples problem when it is applied to face recognition task. In addition, it is only a linear feature extraction technique. In this paper，kernel uncorrelated space algorithm is proposed. The key of the proposed algorithm is to how to compute the uncorrelated space in the higher dimensional feature space. As to this problem, a very simple and easy method is proposed, which originates from the eigenface that transforms the computation of the high order matrix into the computation of the low order matrix, and then the actual computation of the uncorrelated space in the higher dimensional feature space is reduced to a standard eignenvalue problem. In addition, the proposed algorithm can effectively overcome small size samples problem. The numerical experiments on facial databases of ORL show that the proposed method is effective and feasible.

HTML全文

参考文献(1)

Jin Z, Yang J, and Hu Z, et al.. Face recognition based on theuncorrelated discrimination transformation [J]. PatternRecognition, 2001, 34(7): 1405-1416.[2]陈绵书, 陈贺新, 刘伟. 一种新的求解无相关鉴别矢量集方法[J]. 计算机学报, 2004, 27(7): 913-917.Chen Mian-shu, Chen He-xin, and Liu Wei. A new method forresolving the uncorrelated set of discriminant vectors [J].Chinese Journal of Computers, 2004, 27(7): 913-917.[3]Sch.lkopf B.[J].Smola A, and Mller K R. Nonlinear componentanalysis as a kernel eigenvalue problem [R]. Max-Planck-Institut fr biologische Kybernetik, Technical Report.1996,44:-[4]Bach F R and Jordan M I. Kernel independent componentanalysis[R]. Department of Computer Science, University ofCalifornia, Technical Report UCB/CSD-01-1166, 2001.[5]Mika S, Ratsch G, and Weston J, et al.. Fisher discriminantanalysis with kernels [C].Proceedings of IEEE InternationalWorkshop on Neural Networks for Signal Processing.Madison, Wisconsin, August 1999: 41-48.[6]Yang J, Frangi A, and Yang J Y, et al.. KPCA plus LDA: Acomplete kernel Fisher discriminant framework for featureextraction and recognition [J].IEEE Transactions on PatternAnalysis Machine Intelligence.2005, 27(2):230-244[7]梁志贞, 施鹏飞. 核不相关鉴别分析以及它在字符识别中的应用[J]. 计算机研究与发展, 2006, 43(1): 132-137.Liang Zhi-zhen and Shi Peng-fei. Kernel uncorrelateddiscriminant analysis and its application to handwrittencharacter recognition [J]. Journal of Computer Research andDevelopment, 2006, 43(1): 132-137.[8]Turk M and Pentland A. Eigenfaces for recognition[J].Journal of Cognitive Neuroscience.1991, 3(1):71-86[9]Liu K and Yang J Y. An efficient algorithm for Foley-Sammon optimal set of discriminant vectors by algebraicmethod[J].International Journal of Pattern Recognition andArtificial Intelligence.1992, 6(5):817-829[10]Zheng W M and Zhao L. Foley-Sammon optimaldiscriminant vectors using kernel approach [J].IEEETransactions on Neural Networks.2005, 16(1):1-9

施引文献

资源附件(0)

访问统计