期刊文献+

基于图像欧氏距离的二维局部多样性保持投影 被引量:6

Image Euclidean Distance-based Two-dimensional Local Diversity Preserving Projection
下载PDF
导出
摘要 主成分分析可以较好地保持数据的全局多样性几何属性,在模式识别、机器学习、图像识别等领域有着很重要的作用.缺点是他不能较好地保持局部数据的多样性几何属性,且忽略了图像像素之间的相互关系,导致算法性能不够好,且对模式形变比较敏感.对此问题,提出了一种基于图像欧氏距离的二维局部多样性保持投影.该方法利用邻接图描述局部数据之间的变化关系,然后利用图像欧氏距离度量数据间的多样性几何属性,有效地将图像像素之间的相互关系嵌入到目标函数中.和主成分分析相比,所提方法较好地保持了局部数据的多样性几何属性,而且明确考虑了图像像素之间的相互关系,对模式形变具有好的鲁棒性.在Yale,AR及PIE三个人脸库上的实验结果证明了所提算法的有效性. Previous works have demonstrated that principal component analysis(PCA) well preserves the global information,i.e.,diversity of data,and plays an important role in pattern recognition,machine learning,and image processing.However,PCA ignores the spatial relationships among pixels in images and does not well preserve the local diversity of data,which will impair the recognition accuracy and lead to unstableness to the perturbation of images.To address these problems,a novel approach,namely image Euclidean distance based two-dimensional local diversity preserving projection(IED-2DLDPP) is proposed.IED-2DLDPP constructs an adjacency graph to model the variation of data and employs image Euclidean distance to characterize the diversity of data,which explicitly considers the spatial relationships among pixels in the images.Extensive experiments on Yale,AR,and PIE databases show the efficiency of the proposed method.
出处 《自动化学报》 EI CSCD 北大核心 2013年第7期1062-1070,共9页 Acta Automatica Sinica
基金 国家自然科学基金(61271296 60802075) 陕西省自然科学基础研究计划(2012JM8002) 浙江大学CAD&CG国家重点实验室开放课题(A1106) 中国博士后基金(2012M521747) 高等学校学科创新引智计划(B08038) 中央基本科研业务费 西安电子科技大学ISN国家重点实验室自主研究课题资助~~
关键词 二维主成分分析 多样性 流形学习 特征提取 人脸识别 Two-dimensional principal component analysis(2DPCA) diversity manifold learning feature extraction face recognition
  • 相关文献

参考文献3

二级参考文献40

  • 1Yan S C, Xu D, Zhang B, Zhang H, Yang Q, Lin S. Graph embedding and extensions: a general framework for dimensionality reduction. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(1): 40-51. 被引量:1
  • 2Murase H, Nayar S K. Visual learning and recognition of 3-D objects from appearance. International Journal of Computer Vision, 1995, 14(1): 5-24. 被引量:1
  • 3Turk M A, Pentland A P. Face recognition using eigenfaces. In: Proceedings of the Conference on Computer Vision and Pattern Recognition. Hawaii, USA: IEEE, 1991. 586-591. 被引量:1
  • 4Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs fisherfaces: recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7): 711-720. 被引量:1
  • 5Tenenbaum J B, de Silva V, Langford J C. A global geometric framework for nonlinear dimensionality reduction. Science, 2000, 290(5500): 2319-2323. 被引量:1
  • 6Seung H S, Lee D D. The manifold ways of perception. Science, 2000, 290(5500): 2268-2269. 被引量:1
  • 7Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear embedding. Science, 2000, 290(5500): 2323-2326. 被引量:1
  • 8Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and data representation. Neural Computation, 2003, 15(6): 1373-1396. 被引量:1
  • 9He X F, Yan S C, Hu Y, Niyogi P, Zhang H. Face recognition using Laplacianfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(3): 328-240. 被引量:1
  • 10Cai D, He X F, Han J W. Document clustering using locality preserving indexing. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(12): 1624-1637. 被引量:1

共引文献62

同被引文献128

引证文献6

二级引证文献52

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