期刊文献+

一种反对称变换下的有监督局部保持投影算法

Supervised Locality Preserving Projection Based on Anti-symmetric Transformation
下载PDF
导出
摘要 提出一种基于反对称变换下的有监督局部保持投影方法。首先针对监督局部保持投影(Supervised locali-ty preserving projection,SLPP)存在过学习和不能较好地保持图像空间的差异信息等问题,通过最小化局部离散度和最大化差异离散度准则提取投影方向,寻找到SLPP与直接线性鉴别分析(Direct linear discriminant a-nalysis,D-LDA)的一种特殊等价形式。其次,反对称矩阵是一种特殊矩阵,与其相对应的反对称变换是欧氏空间中一类重要的线性变换,本文进一步对特殊等价形式的D-LDA转换矩阵进行反对称变换,得到其对应的反对称矩阵,通过对线性空间下两种矩阵的分别求解,从而得到样本完备的鉴别信息。本文所提方法有效解决了线性空间下小样本问题的特征抽取问题,在NUST603和ORL人脸库上的实验结果验证了该方法的有效性。 Classification of nonlinear high-dirnensional data is usually not amenable to standard pattern recognition techniques because of an underlying small sample size condition. To address this problem, a novel supervised locality preserving projection(SLPP) learning algorithm based on the anti-symmetric transformation is developed. Firstly, according to the problem that SLPP has the over-learning problem and does not preserve the diversity information of data which is also useful for data recognition, a concise transformation of feature extraction criterion is raised by minimizing the local scatter, which efficiently preserves the local structure and simultaneously maximizes the diversity scatter. Furthurmore, a special equivalent form of direct linear discriminant analysis (D- LDA) is obtained. Secondly, anti-symmetric matrix is a kind of specific matrix in matrix theory, and skew symmetric transformation is the basic linear transformation in Euclidean space. By this special equivalent form of D - LDA and anti-sym- metric transformation, two solution spaces derived from the local scatter matrix and its corresponding anti-symmetric subspace are respectively utilized to obtain the efficient discriminatory information of the samples. Therefore, the shortcoming of nonlinear small sample sizes problem in the traditional subspace learning algorithms is overcome. Experimental results on the NUST603 and ORL face databases demonstrate the effectiveness of the proposed method.
出处 《南京航空航天大学学报》 EI CAS CSCD 北大核心 2013年第2期271-276,共6页 Journal of Nanjing University of Aeronautics & Astronautics
基金 国家自然科学基金(61100116)资助项目 江苏省自然科学基金(BK2011371 BK2012700)资助项目 中国博士后科学基金(2011M500926)资助项目 江苏省博士后科学基金(1102063C)资助项目 人工智能四川省重点实验室开放基金重点课题(2012RZY02)资助项目
关键词 特征抽取 局部保持投影 小样本问题 反对称变换 feature extraction locality preserving projection small sample size problem anti-symmetric transformation
  • 相关文献

参考文献15

  • 1孙廷凯,冯爱民,陈松灿.基于相关性度量的伪主成分分析[J].南京航空航天大学学报,2006,38(6):732-736. 被引量:3
  • 2周大可,唐振民.一种加权核鉴别分析方法[J].南京航空航天大学学报,2008,40(2):226-229. 被引量:2
  • 3Tenenbaum J B, de Silva V, Langford J C. A global geometric framework for nonlinear dimensionality reduction[J]. Science,2000,290(5500) :2319-2323.
  • 4Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear embedding [J]. Science, 2000,290 (5500) : 2323-2326.
  • 5Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and data representation[J]. Neural Computation, 2003,15 (6) : 1373-1396.
  • 6He X F, Yan S C, Hu Y, et al. Face recognition using Laplacianfaees[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005,27 (3) : 328-240.
  • 7Cai D, He X F, Han J W. Document clustering using locality preserving indexing[J]. IEEE Transactions on Knowledge and Data Engineering, 2005, 17 (12) : 1624-1637.
  • 8Hu D W, Feng G Y, Zhou Z, Two-dimensional locality preserving projections (2DLPP) with its application to palmprint recognition[J]. Pattern Recognition, 2007,40 (1) : 339-342.
  • 9Yang J, Zhang D, Yang J, et al. Globally maximizing, locally minimizing: unsupervised discriminant projection with applications to face and palm biometrics[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007,29(4) : 650-664.
  • 10Deng W, Hu J, Guo J, et al. Comments on globally maximizing, locally minimizing: unsupervised discriminant projection with application to face and palm biometrics[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(8):1503-1504.

二级参考文献29

  • 1Duda R O,Hart P E,Stock D G.Pattern classification[M].2nd edition.New York:John Wiley & Sons,2000.
  • 2Turk M,Pentland A.Eigenface for recognition[J].Journal of Cognitive Neuroscience,1991,3(1):71-86.
  • 3Wang Liwei,Zhang Yan,Feng Jufu.On the euclidean distance of images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1334-1339.
  • 4Vytautas P.Distance measures for PCA-based face recognition[J].Pattern Recognition Letters,2004,25(6):711-724.
  • 5Chen Songcan,Sun Tingkai.Class-information-incorporated principal component analysis[J].Neurocomputing,2005,69(1-3):216-223.
  • 6Yang Jian,Zhang D,Frangi A F,et al.Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
  • 7Chen Songcan,Zhu Yulian,Zhang Daoqiang,et al.Feature extraction approaches based on matrix pattern:MatPCA and MatFLDA[J].Pattern Recognition Letters,2005,26(8):1157-1167.
  • 8Zhang Daoqiang,Zhou Zhihua.(2D)2PCA:Two-directional two-dimensional PCA for efficient face representation and recognition[J].Neurocomputing,2005,69(1-3):224-231.
  • 9Wang Liwei,Wang Xiao,Zhang Xuerong,et al.The equivalence of two-dimensional PCA to line-based PCA[J].Pattern Recognition Letters,2005,26(1):57-60.
  • 10Martinez A M,Kak A C.PCA versus LDA[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,23(2):228-233.

共引文献36

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部