基于马氏距离的半监督鉴别分析及人脸识别

史骏; 陈才扣

基于马氏距离的半监督鉴别分析及人脸识别

史骏,
陈才扣

扬州大学信息工程学院, 扬州 225009

基金项目: 国家自然科学基金资助项目(60875004); 江苏省高校自然科学基金资助项目(07KJB520133); 江苏省自然科学基金资助项目(BK2009184)

详细信息

中图分类号: TP 391
计量
- 文章访问数: 3306
- HTML全文浏览量: 170
- PDF下载量: 749
- 被引次数: 0
出版历程
- 收稿日期: 2010-07-21
- 网络出版日期: 2012-12-30

Mahalanobis distance-based semi-supervised discriminant analysis for face recognition

Department of Information Engineering, Yangzhou University, Yangzhou 225009, China

摘要

摘要: 针对人脸识别应用中人脸样本的类别信息不足以及人脸样本特征间存在相关性的问题,提出了一种基于马氏距离的半监督鉴别分析.该方法在图嵌入框架下利用马氏距离对数据集中带有类别信息的样本进行边界Fisher分析,不仅保持了类内的紧致性和类间的分离性,而且抽取出有利于分类的鉴别特征,同时将不带类别信息的样本用于描述数据集的几何结构,保留了样本间的局部邻域信息.与传统的特征抽取方法相比,该方法有较好的识别性能,在ORL,YALE及AR人脸数据库上的实验验证了该方法的有效性.
- 特征抽取 /
- 马氏距离 /
- 半监督 /
- 人脸识别
Abstract: To address the problems that there is often no sufficient class-label information of face samples in face recognition application and some relativity also exist among face sample features, a semi-supervised discriminant analysis based on Mahalanobis distance was presented. The method makes use of the Mahalanobis distance to perform marginal fisher analysis (MFA) for labeled samples in the data set, which is on the basis of the graph embedding framework, so that it not only preserves the intraclass compactness and the interclass separability, but also extracts the discriminant characteristics for effective classification, and simultaneously the unlabeled samples were utilized to characterize the geometric structure of the data set, and thus the local neighborhood information among samples was well preserved. Compared with the traditional feature extraction methods, the proposed method has better recognition performance, and the experiments on ORL, YALE and AR face databases demonstrate the effectiveness of this method.
- feature extraction /
- Mahalanobis distance /
- semi-supervised /
- face recognition

HTML全文

参考文献(1)

[1] Turk M A,Pentland A P.Face recognition using eigenfaces //Proc 91 IEEE Comput Soc Conf Comput Vision Pattern Recognit.Piscataway,NJ:IEEE,1991:586-591 [2] Belhumeur P N,Hespanha J P,Kriegman D J.Eigenfaces vs Fisherfaces:recognition using class specific linear projection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(7):711-720 [3] He Xiaofei,Niyogi P.Locality preserving projections //Proceedings of the Conference on Advances in Neural Information Processing Systems.Cambridge:MIT Press,2003:153-160 [4] Yan Shuicheng,Xu Dong,Zhang Benyu,et al.Graph embedding and extension:a general framework for dimensionality reduction[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(1):40-51 [5] Xing E P,Ng A Y,Jordan M I,et al.Distance metric learning,with application to clustering with side-information[M].Cambridge:MIT Press,2003:505-512 [6] 何晓群.多元统计分析[M].北京:中国人民大学出版社,2004:6-76 He Xiaoqun.Multivariate statistical analysis[M].Beijing:Renmin University of China Press,2004:6-76(in Chinese) [7] 黄飞,周军,卢晓东.基于马氏距离的一维距离像识别算法仿真[J].计算机仿真,2010,27(3):31-34 Huang Fei,Zhou Jun,Lu Xiaodong.The simulation of one-dimensional range profile recognition based on mahalanobis distance[J].Computer Simulation,2010,27(3):31-34(in Chinese) [8] Aouada D,Baryshnikov Y,Krim H.Mahalanobis-based adaptive nonlinear dimension reduction //Proceedings-International Conference on Pattern Recognition.Piscataway,NJ:IEEE,2010:742-745 [9] Cai Deng,He Xiaofei,Han Jiawei.Semi-supervised discriminant analysis //Proceedings of the IEEE International Conference on Computer Vision.Piscataway,NJ:IEEE,2007:1-7 [10] Remus S,Tomasi C.Semi-supervised fisher linear discriminant (SFLD) //IEEE International Conference on Acoustics,Speech and Signal Processing.Piscataway,NJ:IEEE,2010:1862-1865

施引文献

资源附件(0)

访问统计