[1]金一,阮秋琦.一种局部加权的二维主成分分析算法及其在人脸识别中的应用[J].智能系统学报,2007,2(03):25-29.
 JIN Yi,RUAN Qiu-qi.A partially weighted twodimensional PCA for face recognition[J].CAAI Transactions on Intelligent Systems,2007,2(03):25-29.
点击复制

一种局部加权的二维主成分分析算法及其在人脸识别中的应用(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第2卷
期数:
2007年03期
页码:
25-29
栏目:
出版日期:
2007-06-25

文章信息/Info

Title:
A partially weighted twodimensional PCA for face recognition
文章编号:
1673-4785(2007)03-0025-05
作者:
金一 阮秋琦
北京交通大学计算机与信息技术学院,北京100044
Author(s):
JIN Yi RUAN Qiu-qi
College of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
关键词:
二维主成分分析局部加权人脸识别加权特征提取
Keywords:
twodimensional principal component analysis partiallyweighted face recognition weighted feature extraction
分类号:
TP391.4
文献标志码:
A
摘要:
提出了一种将局部特征加权与二维主成分分析相结合的局部加权的二维主成分分析方法.引入了二维局部加权特征子空间的概念,将各类样本映射到这个局部加权特征子空间,再通过计算测试样本到加权子空间的距离进行样本的分类.使用这种方法在ORL人脸库上进行测试,结果表明,经过局部特征加权的二维主成分分析方法比普通的二维主成分分析方法具有更优的性能,并且在提高识别率的同时算法的复杂程度并没有明显增加.
Abstract:
This paper proposes face recognition software that uses twodimensional principal component analysis (2DPCA) in conjunction with partial feature weighting by applying twodimensional partialweighting to the characteristic subspace. First faces are mapped onto this partially weighted 2DPCA subspace, then the samples are classified by calculating the distance from the samples to the partially weighted 2DPCA subspace. To test this new method, ORL face databases were used and it was found that the recognition rate was higher than with either 2DPCA or PCA and the computational complexity did not increase significantly.

参考文献/References:

[1] KIRBY M, SIROVICH L. Application of the Karhunen Loeve procedure for the ch aracterization of human faces[J]. IEEE Transactions on Pattern Analysis and Ma chine Intelligence, 1990, 12 (1): 103-108.
[2]TURK M, PENTLAND A. Eigenfaces for recognition[J]. Journal of Cogn itive Neuroscience, 1991, 3 (1): 72-86.
[3]BELHUMEUR P N, HESPANHA J P, KRIEGMAN D J. Eigenfaces vs. Fisherface : Recog nition using class special linear projection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19 (7): 711-720.
【4]BARTLETT. Face recognition by independent component analysis[ J]. I EEE Transactions on Neural Networks, 2002,13(6): 1450-1464.
[5]刘青山.综述人脸识别中的子空间方法[J].自动化学报,2003, 129(16):900 -912.
LIU Qingshan. A survey: subspace analysis for face recognition[J]. Acta Automa tica Sinica, 2003, 129(16):900-912.
[6]YANG J, ZHANG D, YANG J Y. Twodimensional PCA: A new approach to a pp earance based face representation and recognition [J].IEEE Transactions Pattern Anal Machine Intell, 2004, 26 (1): 131-137.
[7]乔 宇,黄席樾,柴 毅. 基于加权主元分析(WPCA)的人脸识别[J ]. 重庆大学学报,2004, 27(3):28-31. 
 QIAO Yu, HUANG Xiyue, CHAI Yi.Face recognition based weighted PCA[J]. Journal of Chongqing University, 2004, 27(3):28-31.
[8]杨 光,阮秋琦.一种新的基于加权主分量分析的人脸识别算法[A].中国人工智能学会第11届全国学术年会(CAAI-11)[C] .武汉,中国,2005.
YANG Guang, RUAN Qiuqi. A new arithmetic of face recognition based on weighted P CA[A]. The 11th China Artificial Intelligence Conference[C].Wuhan, China,200 5.

相似文献/References:

[1]王科俊,贲晛烨,刘丽丽.采用Radon变换和二维主成分分析的步态识别算法[J].智能系统学报,2010,5(03):266.
 WANG Ke-jun,BEN Xian-ye,LIU Li-li.Gait recognition with Radon transform and D principal component analysis[J].CAAI Transactions on Intelligent Systems,2010,5(03):266.
[2]胡娜,马慧,湛涛.融合LBP纹理特征与B2DPCA技术的手指静脉识别方法[J].智能系统学报,2019,14(03):533.[doi:10.11992/tis.201801014]
 HU Na,MA Hui,ZHAN Tao.Finger vein recognition method combining LBP texture feature and B2DPCA technology[J].CAAI Transactions on Intelligent Systems,2019,14(03):533.[doi:10.11992/tis.201801014]

备注/Memo

备注/Memo:
收稿日期:2006-09-08.
基金项目:国家重点基础研究发展计划(973)资助项目(2004BC318005);教育部博士点基金资助项目(60 472033)
作者简介:
金 一,女,1982年生,博士研究生,主要研究方向为图像处理、模式识别、人脸识别. E-mail:jineejin@gmail.com.
阮秋琦,男,1944年生,教授,博士生导师,主要研究方向为图像处理、计算机视觉、模式识别、虚拟现实.曾多次获得省部级科技进步奖,发表论文150余篇,出版专著3部.
更新日期/Last Update: 2009-05-06