首页 | 本学科首页   官方微博 | 高级检索  
     

融合复小波与零空间判别投射的人脸识别方法
引用本文:张强. 融合复小波与零空间判别投射的人脸识别方法[J]. 光电子.激光, 2009, 20(9): 1208-1213
作者姓名:张强
作者单位:上海交通大学电子信息与电气工程学院,上海,200240;上海交通大学电子信息与电气工程学院,上海,200240;上海理工大学,上海,200093;上海系统科学研究院,上海,200093
基金项目:国家“973”计划资助项目(2002cb312200); 国家自然科学基金资助项目(60575036)
摘    要:提出一种新颖的零空间判别投射(NDPE)的子空间人脸识别方法。基于局部保持映射(LPP)和非参数判别分析方法,NDPF能够同时编码人脸数据流形的几何和判别结构,并且通过在零空间中求解特征值来克服小样本尺寸问题。为进一步提高人脸识别的准确率,提出融合双树复小波变换(DTCWT)与NDPE的方法。实验结果表明,所提人脸识别方法在ORL、Yale和AR人脸数据库上均取得了较高的识别率。

关 键 词:人脸识别  零空间判别投射(NDPE)  双树复小波变换(DTCWT)  特征融合

An approach for face recognition based on fusion of complex wavelet and null space discriminant projection embedding
ZHANG Qiang. An approach for face recognition based on fusion of complex wavelet and null space discriminant projection embedding[J]. Journal of Optoelectronics·laser, 2009, 20(9): 1208-1213
Authors:ZHANG Qiang
Affiliation:ZHANG Qiang1,CAI Yun-ze1,XU Xiao-ming1,2,3 (1.School of Electric and Information Engineering,Shanghai Jiaotong University,Shanghai 200240,China,2.University of Shanghai for Science and Technology,Shanghai 200093,3.Shanghai Academy of Systems Science,China)
Abstract:A novel face recognition method called null space discriminant projection embedding(NDPE) is proposed.Based on locality preserving projections(LPP) and nonparametric discriminant analysis methods,NDPE can encode both the geometrical and discriminant structure of the face data manifold,and address the small sample size problem by solving an eigenvalue problem in null space.Dual tree complex wavelet transformation(DTCWT) is integrated with NDPE to improve the recognition performance.Experiment results on ORL,...
Keywords:face recognition  null space discriminant projection embedding(NDPE)  dual tree complex wavelet transformation(DTCWT)  feature fusion  
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《光电子.激光》浏览原始摘要信息
点击此处可从《光电子.激光》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号