[1]李晓东,费树岷,张涛.基于差空间和最大散度差鉴别分析的人脸识别方法[J].东南大学学报(自然科学版),2009,39(6):1130-1134.[doi:10.3969/j.issn.1001-0505.2009.06.009]
 Li Xiaodong,Fei Shumin,Zhang Tao.Face recognition based on residual space and maximum scatter difference[J].Journal of Southeast University (Natural Science Edition),2009,39(6):1130-1134.[doi:10.3969/j.issn.1001-0505.2009.06.009]
点击复制

基于差空间和最大散度差鉴别分析的人脸识别方法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
39
期数:
2009年第6期
页码:
1130-1134
栏目:
计算机科学与工程
出版日期:
2009-11-20

文章信息/Info

Title:
Face recognition based on residual space and maximum scatter difference
作者:
李晓东12 费树岷1 张涛1
1 东南大学复杂工程系统测量与控制教育部重点实验室, 南京 210096; 2 临沂师范学院信息学院, 临沂 276005
Author(s):
Li Xiaodong12 Fei Shumin1 Zhang Tao1
1 Key Laboratory of Measurement and Control of Complex Systems of Engineering of Ministry of Education, Southeast University, Nanjing 210096,China
2 School of Information, Linyi Normal College, Linyi 276005,China
关键词:
二维主成份分析 类内平均脸 差空间 最大散度差鉴别分析
Keywords:
two dimension principal component analysis(2DPCA) with-in class average face residual space maximum scatter difference discriminate analysis
分类号:
TP391.41
DOI:
10.3969/j.issn.1001-0505.2009.06.009
摘要:
为了提高最大散度差鉴别分析方法在人脸识别中的识别率,提出了一种改进的基于差空间的最大散度差鉴别分析人脸识别算法.该方法把类内平均脸方法应用到2DPCA算法中,并基于改进的2DPCA方法分别建立训练样本和测试样本的差空间,然后用类内中间值代替类内均值修改了最大散度差鉴别算法中类内散布矩阵的定义.用改进后的最大散度差鉴别法对得到的差空间进行鉴别分析,分别提取训练样本和测试样本的鉴别特征,用最近邻分类器分类.在ORL人脸数据库上的实验结果表明,该方法可以有效地改善识别率.
Abstract:
To improve the recognition rate of maximum scatter difference(MSD), A modified method of discriminate feature extraction based on maximum scatter difference criterion in residual space is proposed. Firstly, within-class average face is combined with two dimension principal component analysis(2DPCA). The improved 2DPCA is used to construct residual spaces of training samples and testing samples. At the same time, the definition of within-class matrix which is in the definition of MSD is modified by replacing within-class mean vector with within-class median vector. Then improved maximum scatter difference discriminate analysis is performed on the residual space to extract discriminate features of training samples and testing samples. Finally, nearest distance classifier is conducted for classifying. A lot of experiments results based on ORL(Olivetti research laboratory)face database show that the proposed algorithm can improve the recognition rate.

参考文献/References:

[1] Zhao W,Chellappa R,Phillips P J,et al.Face recognition:a literature survey[J].Acm Computing Surveys,2003,35(4):399-459.
[2] 刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法[J].自动化学报,2003,29(6):900-911.
  Liu Qingshan,Lu Hanqing,Ma Songde.A survey:subspace analysis for face recognition[J].Acta Automatica Sinica,2003,29(6):900-911.(in Chinese)
[3] Turk M,Pentland A.Eigenfaces for recognition[J].Cognitive Neuroscience,1991,3(1):71-86.
[4] Lu J,Plataniotis K,Venetsanopoulos A.Face recognition using LDA-based algorithms[J].IEEE Trans Neural Networks,2003,14(1):195-200.
[5] Kwak K C,Pedrycz W.Face recognition using an enhanced independent component analysis approach[J].IEEE Trans Neural Networks,2007,18(2):530-541.
[6] Yang J,Zhang D.Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[7] Kim H C,Kim D,Bang S Y,et al.Face recognition using the second-order mixture-of-eigenfaces method[J].Pattern Recognition,2004,37(2):337-349.
[8] 刘永俊,陈才扣.基于差空间的最大散度差鉴别分析及人脸识别[J].计算机应用,2006,26(10):2460-2465.
  Liu Yongjun,Chen Caikou.Maximum scatter difference discriminant analysis in residual space and face recognition[J].Journal of Computer Applications,2006,26(10):2460-2465.(in Chinese)
[9] Yang Jian,Zhang David,Frangi Alejandro F,et al.Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[10] 何国辉,甘俊英.PCA类内平均脸法在人脸识别中的应用研究[J].计算机应用研究,2006,23(3):165-169.
  He Guohui,Gan Junying.Study for within-class average face method based on PCA in face recognition [J].Application Research of Computers,2006,23(3):165-169.(in Chinese)
[11] Li Haifeng,Jiang Tao,Zhang Keshu.Efficient and robust feature extraction by maximum margin criterion [J].IEEE trans on Neural Network,2006,17(1):157-165.
[12] 宋枫溪,程科,杨静宇,等.最大散度差和大间距线性投影与支持向量机[J].自动化学报,2004,30(6):890-896.
  Song Fengxi,Cheng Ke,Yang Jingyu,et al.Maximum scatter difference,large margin linear projection and support vector machines[J]. Acta Automatica Sinica,2004,30(6):890-896.(in Chinese)
[13] Yang J,Zhang D,Yang J Y.Median LDA:a robust feature extraction method for face recognition[C] //Proc of IEEE International Conference on Systems,Man,and Cybernetics.Taiwan,China,2006:4208-4213.

备注/Memo

备注/Memo:
作者简介: 李晓东(1974—),男,博士生; 费树岷(联系人),男,博士,教授,博士生导师,smfei@seu.edu.cn.
基金项目: 国家自然科学基金资助项目(60574006).
引文格式: 李晓东,费树岷,张涛.基于差空间和最大散度差鉴别分析的人脸识别方法[J].东南大学学报:自然科学版,2009,39(6):1130-1134. [doi:10.3969/j.issn.1001-0505.2009.06.009]
更新日期/Last Update: 2009-11-20