[1]王洪元,刘爱萍,冯燕.一种改进的ISOMAP算法在图像检索中的应用[J].常州大学学报(自然科学版),2011,(04):41-44.
　WANG Hong-yuan,LIU Ai-ping,FENG Yan.An Improved Kernel ISOMAP Algorithm with Application to Image Retrieval[J].Journal of Changzhou University(Natural Science Edition),2011,(04):41-44.
点击复制

一种改进的ISOMAP算法在图像检索中的应用()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
期数:
2011年04期
页码:
41-44
栏目:
出版日期:
2011-09-30

文章信息/Info

Title:
An Improved Kernel ISOMAP Algorithm with Application to Image Retrieval
作者:
王洪元1; 2; 刘爱萍1; 2; 冯燕1; 2
1.常州大学 信息科学与工程学院, 江苏 常州 213164；2.常州市过程感知与互联技术重点实验室,江苏 常州 213164
Author(s):
　WANG Hong-yuan1; 2;  LIU Ai-ping1; 2;  FENG Yan1; 2
1School of Information Science and Engineering, Changzhou University，Changzhou 213164, China； 2Changzhou Key Laboratory for Process Perception and Interconnected Technology，Changzhou 213164, China
关键词:
非线性降维; 核等距特征映射; 多类多流形等距特征映射; 图像检索
Keywords:
nonlinear dimensionality;  kernel ISOMAP;  multi-class multi-manifold ISOMAP;  image retrieval
分类号:
TP 391.43
文献标志码:
A
摘要:
传统的核化ISOMAP（K-ISOMAP）算法对于多个分散类簇数据集的低维映射不能较好地表现数据集的内在拓扑结构。针对此缺点，本文将对基于ISOMAP的多类多流形算法（MCMM-ISOMAP）进行核化，提出核化的多类多流形ISOMAP算法（K-MCMM-ISOMAP），该算法不仅使得多类数据集在降维后保持较好的内在拓扑结构，而且具备了K-ISOMAP算法的泛化能力，可以将测试数据直接映射到低维空间。因此，该算法可以在多类图像数据集中实现图像检索的功能。实验结果表明该算法与K-ISOMAP相比更具有效性。
Abstract:
The conventional kernel ISOMAP algorithm(K-ISOMAP) can not work well in keeping the intrinsic topology of datasets from multi-class clusters datasets in the low-dimensional space. In order to avoid this shortcoming, a novel algorithm named kernel multi-class multi-manifold ISOMAP (K-MCMM-ISOMAP) is proposed in this paper, which is the kernel version of MCMM-ISOMAP. The new algorithm doesn't only keep the intrinsic topology of datasets in low-dimensional mapping space, but also has the generalization of K-ISOMAP. It can directly map the test data to low-dimensional space. Therefore it can be applied to the image retrieval system consisting of multi-class image dataset. The experimental results show that the new algorithm is more effective than the K-ISOMAP.

参考文献/References:

[1]Jolliffe I T. Principal Component Analysis［M］. New York:Springer-Verlag, 1986.
[2］Tenenbaum J B, Silva V de, Langford J C. A global geometric framework for nonlinear dimensionality reduction［J］. Science, 2000,290(12):2 319-2 323.
[3］ Roweis S, Saul L. Nonlinear dimensionality analysis by locally linear embedding［J］. Science, 2000,290(5500):2 323-2 326.
[4］Seung H S ,Lee D D.The maniflold ways of perceotion［J］. Science, 2000, 290(5500):2 268-2 269.
[5］Choi H, Choi S. Kernel Isomap［J］. Electronics Letters,2005,40 (25):1 612-1 613.
[6］ 程起才,王洪元,刘爱萍,等. 基于ISOMAP的一种多流行学习算法［J］.微电子学与计算机,2009,26 (10),115-117.
[7］Nene S A, Nayar S K, Murase H. Columbia automated vision environment:columbia university image library (coil-20)［EB/OL］. ( 2010-09-11 )［ 2011-08-04 ］.http://www1.cs.columbia.edu/CAVE/.1996

备注/Memo

备注/Memo:
基金项目：国家自然科学基金项目（61070121,60973094 ）；江苏省自然科学基金项目（BK2009538）；江苏省产学研前瞻性联合研究项目(BY2009117) 作者简介：王洪元（1962—），男，江苏常熟人，教授，博士，主要从事模式识别与智能系统研究。
更新日期/Last Update: 2011-09-30