| 基于聚类的核主成分分析在特征提取中的应用 |
| |
| 引用本文： | 王和勇,姚正安,李磊.基于聚类的核主成分分析在特征提取中的应用[J].计算机科学,2005,32(4):64-66. |
| |
| 作者姓名： | 王和勇 姚正安 李磊 |
| |
| 作者单位： | 1. 中山大学软件研究所,广州,510275
2. 中山大学数学与计算科学学院,广州,510275 |
| |
| 基金项目： | 国家自然科学基金(10171113) |
| |
| 摘 要： | 本文分析了一般主成分分析在处理非线性问题上的不足,阐述了核主成分分析方法及其计算速度的缺陷,提出了基于聚类的核主成分分析方法。试验结果显示:基于聚类的核主成分分析方法具有好的特征提取性能,相比核主成分分析大大提高了特征提取的速度。
|
| 关 键 词： | 图像检索 KPCA 特征提取 |
The Application of Feature Extraction on Using Kernel Principal Component Analysis Based on Clustering |
| |
| WANG He-Yong,YAO Zheng-An,LI Lei.The Application of Feature Extraction on Using Kernel Principal Component Analysis Based on Clustering[J].Computer Science,2005,32(4):64-66. |
| |
| Authors: | WANG He-Yong YAO Zheng-An LI Lei |
| |
| Affiliation: | WANG He-Yong,YAO Zheng-An,LI Lei Institute of Software,Zhongshan University,Guangzhou 510275 College of Mathematics and Computer Science,Zhongshan University,Guangzhou 510275 |
| |
| Abstract: | This paper points out the drawbacks of the general principal component analysis(PCA) when it is used to solve nonlinear problem firstly. The kernel principal component analysis(KPCA) and its drawbacks on computing are explained secondly. KPCA based on clustering is introduced in the end. The research result shows that the KPCA based on clustering has excellent performance of feature extraction. |
| |
| Keywords: | Image retrieval KPCA KPCA based on clustering Feature extraction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
点击此处可从《计算机科学》下载全文 |
|
