基于非负矩阵分解新的人脸识别方法.pdf

第20 卷第1 期 系 统 仿 真 学 报© Vol. 20 No. 1 2008 年1 月 Journal of System Simulation Jan., 2008 基于非负矩阵分解新的人脸识别方法 1,2 2 李勇智 ，杨静宇 (1.南京林业大学信息科学技术学院, 南京 210037; 2.南京理工大学计算机系, 南京210094) 摘 要：非负矩阵分解是一个新的特征提取方法，基于非矩阵分解的理论，提出了具有正交性的投 影轴的计算方法和具有统计不相关性的投影轴的计算方法。与原非负矩阵分解方法，提出的方法在 某种程度上是降低了特征矢量之间的统计相关性，并且提高识别率。通过在ORL 人脸库和YALE 人脸库上进行实验，结果表明提出的两种特征提取方法在识别率方面整体上好于原非负矩阵分解特 征提取(NMF)方法，甚至超过主成分分析(PCA)法。 关键词：非负矩阵分解；正交投影轴；统计不相关性；特征提取；人脸识别 中图分类号：TP391 文献标识码：A 文章编号：1004-731X (2008) 01-0111-06 Novel Methods of Face Recognition Based on Non-negative Matrix Factorization LI Yong-zhi 1,2, YANG Jing-yu2 (1. School of Information Science Technology, Nanjing Forestry University, Nanjing 210037, China; 2. Department of Computer Science, Nanjing University of Science Technology, Nanjing 210094, China) Abstract: Non-negative matrix factorization (NMF) is a new feature extraction method. Based on the Non-negative matrix factorization (NMF), a new algorithm of orthogonal projection axis and a new algorithm of statistically uncorrelated projection axis for feature extraction were proposed. Compared with original NMF method, the proposed methods are better in terms of reducing or eliminating the statistical correlation between features and improving recognition rate. The experimental results on Olivetti Research Laboratory (ORL) face database and YALE face database show that the new methods are better than original NMF in terms of r