计算机科学 ›› 2018, Vol. 45 ›› Issue (2): 90-93.doi: 10.11896/j.issn.1002-137X.2018.02.015

• 2017年中国计算机学会人工智能会议 • 上一篇    下一篇

广义的鉴别局部中值保持投影及人脸识别

张永,万鸣华   

  1. 南昌航空大学信息工程学院 南昌330063,南昌航空大学信息工程学院 南昌330063;南京审计大学工学院 南京211815
  • 出版日期:2018-02-15 发布日期:2018-11-13
  • 基金资助:
    本文受国家自然科学基金项目(61462064),中国博士后基金项目(2016M600674),江苏省自然科学基金面上项目(BK20161580)资助

Generalized Discriminant Local Median Preserving Projections and Face Recognition

ZHANG Yong and WAN Ming-hua   

  • Online:2018-02-15 Published:2018-11-13

摘要: 针对鉴别的局部中值保持投影(DLMPP)在小样本情况下面临的类内散布矩阵奇异的问题,提出了广义的鉴别局部中值保持投影(GDLMPP)算法。GDLMPP首先将样本等价映射到一个低维子空间,然后在此子空间求解最佳投影矩阵,从而有效解决了小样本问题,并从理论上验证了当类内散布矩阵非奇异时,GDLMPP等价于DLMPP。最后,通过在ORL及AR库上的实验验证了算法的有效性。

关键词: 人脸识别,特征提取,小样本问题,鉴别的局部中值保持投影

Abstract: To solve the problem of the singularity of the within-class scatter matrix in discriminant local median preserving projections (DLMPP) in the case of small sample problem,an algorithm named generalized local median preserving projection (GDLMPP) was proposed.To solve the small sample problem,GDLMPP firstly transforms the samples into a lower dimensional space equivalently,and then solves the optimal projection matrix.The theoretical analysis shows that GDLMPP is equivalent to DLMPP when the within-class scatter matrix is non-singular.At last,the experimental results validate the effectiveness of the proposed algorithm on the ORL and AR face databases.

Key words: Face recognition,Feature extraction,Small sample problem,Discriminant local median preserving projections

[1] TURK M,PENTLAND A.Eigenfaces for recognition [J].Journal of Cognitive Neuroscience,1991,3(1):71-86.
[2] VIDAL R,MA Y,SASTRY S S.Robust Principal ComponentAnalysis[M]∥Generalized Principal Component Analysis.Springer,New York,2016:63-122.
[3] 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.
[4] WANG S,LU J,GU X,et al.Semi-supervised linear discrimi-nant analysis for dimension reduction and classification[J].Pattern Recognition,2016,57(C):179-189.
[5] HE X F,YAN S C,HU Y X,et al.Face recognition using Laplacianfaces[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(3):328-340.
[6] WEN Y,YANG S,HOU L,et al.Face recognition using locality sparsity preserving projections[C]∥2016 International Joint Conference on Neural Networks (IJCNN).IEEE,2016:3600-3607.
[7] YU W W,TENG X L,LIU C Q.Face recognition using discrimi-nant locality preserving projections [J].Image and Vision Computing,2006,24(3):239-248.
[8] HUANG P,TANG Z M.Discriminant of Local Median Preserving Projection with its Application to Face Recognition [J].Journal of Computer-Aided Design & Computer Graphics,2012,24(11):1420-1425.(in Chinese) 黄璞,唐振民.鉴别的局部中值保持投影及其在人脸识别中的应用[J].计算机辅助设计与图形学学报,2012,24(11):1420-1425.
[9] WAN M,LI M,YANG G W,et al.Feature extraction usingtwo-dimensional maximum embedding difference [J].Information Sciences,2014,4(274):55-69.
[10] LAI Z H,WONG W K,XU Y,et al.Approximate orthogonal sparse embedding for dimensionality reduction[J].IEEE Tran-sactions on Neural Networks and Learning Systems,2016,27(4):723-735.
[11] NING X,LI W J,LI H G,et al.Uncorrelated Local Preserving Discriminant Analysis Based on Bionics[J].Journal of Computer Research and Development,2016,53(11):2623-2629.(in Chinese) 宁欣,李卫军,李浩光,等.基于仿生学的不相关局部保持鉴别分析[J].计算机研究与发展,2016,53(11):2623-2629.
[12] MA X H,TAN Y Q.Face recognition based on Discriminantsparse preserving embedding[J].Acta Automatica Sinica,2014,40(1):73-82.(in Chinese) 马小虎,谭延琪.基于鉴别稀疏保持嵌入的人脸识别算法[J].自动化学报,2014,40(1):73-82.
[13] ZHAO Z H,HAO X H.Linear Locality Preserving and Discrimi-nating Projection for Face Recognition [J].Journal of Electroni-cs & Information Technology,2013,35(2):463-467.(in Chinese) 赵振华,郝晓弘.局部保持鉴别投影及其在人脸识别中的应用[J].电子与信息学报,2013,35(2):463-467.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!