计算机科学 ›› 2016, Vol. 43 ›› Issue (1): 35-39.doi: 10.11896/j.issn.1002-137X.2016.01.008

• CRSSC-CWI-CGrC2015 • 上一篇    下一篇

核典型相关分析特征融合方法及应用

许洁,梁久祯,吴秦,李敏   

  1. 江南大学物联网工程学院智能系统与网络计算研究所 无锡214122,江南大学物联网工程学院智能系统与网络计算研究所 无锡214122,江南大学物联网工程学院智能系统与网络计算研究所 无锡214122,江南大学物联网工程学院智能系统与网络计算研究所 无锡214122
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受国家自然科学基金项目(61170121,61202312)资助

Kernel Canonical Correlation Analysis Feature Fusion Method and Application

XU Jie, LIANG Jiu-zhen, WU Qin and LI Min   

  • Online:2018-12-01 Published:2018-12-01

摘要: 构建了一种基于核函数的典型相关分析的特征融合算法。首先,利用核函数将图像矩阵映射到核空间,再抽取同一模式的两组特征向量,在两组特征向量之间建立描述它们的相关性的判据准则函数;然后依此准则函数抽取两组典型投影矢量集;最后通过给定的特征融合策略抽取组合的典型相关特征以用于分类识别。该算法将两组特征向量之间的相关性特征作为有效鉴别信息,既可以很好地融合信息,又可以有效地去除特征之间的信息冗余,并且避免了对映射后的数据矩阵进行分解,从而简化了数据运算。在AR、PIE、ORL、Yale人脸数据库及UCI手写体数字库上的实验结果证明了该方法的有效性和稳定性。

关键词: 核函数,核典型相关分析,特征融合,组合特征抽取,人脸识别

Abstract: A feature fusing algorithm based on kernel canonical correlation analysis KCCA) was constructed in this paper.First,the image data are mapped through kernel function,and then two groups of feature vectors with the same pattern are extracted and the correlation criterion function between the two groups of feature vectors are established.Se-condly,two groups of canonical projective vectors are extracted according to this function.Thirdly,feature fusion for classification is done by using proposed strategy.The advantage of the proposed algorithm lies in the following aspects.Firstly,it suits for information fusion.Secondly,it eliminates the redundant information within the features,and it simplifies the computation without decomposing the mapped matrix and gains more discriminated features.The results of experiments on AR face database,PIE face database,ORL face database,Yale face database and UCI handwritten digit database show that this algorithm is efficient and robust.

Key words: Kernel function,Kernel canonical correlation analysis,Feature fusion,Combined feature extraction,Face recognition

[1] Liu C,Wechsler H.A shape-and texture-based enhanced Fisher classifier for face recognition[J].IEEE Transactions on Image Processing,2001,10(4):598-608
[2] Yang J,Yang J.Generalized K-L transform based combined feature extraction[J].Pattern Recognition,2002,35(1):295-297
[3] Yang J,Yang J,Zhang D,et al.Feature fusion:parallel strategy vs.serial strategy[J].Pattern Recognition,2003,36(6):1369-1381
[4] Fukunaga K.Introduction to statistical pattern recognition[M].Academic press,2013
[5] Worley B,Halouska S,Powers R.Utilities for quantifying separation in PCA/PLS-DA scores plots[J].Analytical biochemistry,2013,433(2):102-104
[6] Mercer T G,Frostick L E.Leaching characteristics of CCA-treated wood waste:a UK study[J].Science of the Total Environment,2012,427:165-174
[7] Hu G D,Zhang R C.Multivariate data analysis methods:Purely algebraic treatment[M].Tianjin:Nankai University Press,1990(in Chinese)胡国定,张润楚.多元数据分析方法:纯代数处理[M].天津:南开大学出版社,1990
[8] Protásio T D P,Trugilho P F,Neves T A,et al.Canonical correlation analysis between characteristics of Eucalyptus wood and charcoal[J].Scientia Forestalis,2012,40(95):317-326
[9] Hong S,Chen X,Jin L,et al.Canonical correlation analysis for RNA-seq co-expression networks[J].Nucleic acids research,2013,41(8):e95
[10] Lin D,Zhang J,Li J,et al.Group sparse canonical correlationanalysis for genomic data integration[J].BMC bioinformatics,2013,14(1):245
[11] Chang P C,Wu J L.A critical feature extraction by kernel PCA in stock trading model[J].Soft Computing, 2015,19(5):1393-1408
[12] Lai P L,Fyfe C.Kernel and nonlinear canonical correlation ana-lysis[J].International Journal of Neural Systems,2000,10(5):365-377
[13] Akaho S.A kernel method for canonical correlation analysis[C]∥Proceedings of the International Meeting of the Psychometric Society(MPS 2001).2001
[14] Joachims T,Cristianini N,Shawe-Taylor J.Composite kernelsfor hypertext categorisation[C]∥Proceeding of the Eighteenth International Conference on Machine Learning.Morgan Kaufmann Publishers Inc.,2001:250-257
[15] Knutsson H,Borga M.Learning multidimensional signal pro-cessing[D].Linkping University,1998
[16] Hotelling H.Relations between two sets of variates[J].Bio-metrika,1936,28(314):321-377
[17] Sun Q S.Research on feature extraction and image recognition based on correlation Projection analysis[D].Nanjing:Nanjing University of Science and Technology,2006(in Chinese)孙权森.基于相关投影分析的特征抽取与图像识别研究[D].南京:南京理工大学,2006

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 雷丽晖,王静. 可能性测度下的LTL模型检测并行化研究[J]. 计算机科学, 2018, 45(4): 71 -75, 88 .
[2] 厉柏伸,李领治,孙涌,朱艳琴. 基于伪梯度提升决策树的内网防御算法[J]. 计算机科学, 2018, 45(4): 157 -162 .
[3] 王欢,张云峰,张艳. 一种基于CFDs规则的修复序列快速判定方法[J]. 计算机科学, 2018, 45(3): 311 -316 .
[4] 孙启,金燕,何琨,徐凌轩. 用于求解混合车辆路径问题的混合进化算法[J]. 计算机科学, 2018, 45(4): 76 -82 .
[5] 伍建辉,黄中祥,李武,吴健辉,彭鑫,张生. 城市道路建设时序决策的鲁棒优化[J]. 计算机科学, 2018, 45(4): 89 -93 .
[6] 钟菲,杨斌. 基于主成分分析网络的车牌检测方法[J]. 计算机科学, 2018, 45(3): 268 -273 .
[7] 史雯隽,武继刚,罗裕春. 针对移动云计算任务迁移的快速高效调度算法[J]. 计算机科学, 2018, 45(4): 94 -99, 116 .
[8] 王帅,刘娟,毕姚姚,陈哲,郑群花,段慧芳. 基于两步聚类和随机森林的乳腺腺管自动识别方法[J]. 计算机科学, 2018, 45(3): 247 -252 .
[9] 耿海军,施新刚,王之梁,尹霞,尹少平. 基于有向无环图的互联网域内节能路由算法[J]. 计算机科学, 2018, 45(4): 112 -116 .
[10] 李珊,饶文碧. 基于视频的矿井中人体运动区域检测[J]. 计算机科学, 2018, 45(4): 291 -295 .