Share Email Print

Proceedings Paper

A mixture feature selection method for remote sensing image
Author(s): Xiaochun Cai; Yihua Hu; Xiaohong Tao; Guilan Hu
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The need for remote sensing image feature selection methods is discussed in this paper. A central problem in image classification and recognition is the redundancy of image features. To cope with many unnecessary and irrelevant features, we propose a mixture method based on principle component analysis (PCA) and rough set theory to alleviate this situation. The main contribution of this paper is to provide the method for remote sensing image classification with higher accuracy comparing to the single rough set theory and PCA method. Finally, some experimental results demonstrate that our proposed method is effective in feature selection for remote sensing image.

Paper Details

Date Published: 4 January 2006
PDF: 5 pages
Proc. SPIE 5985, International Conference on Space Information Technology, 598530 (4 January 2006); doi: 10.1117/12.657920
Show Author Affiliations
Xiaochun Cai, Electronic Engineering Institute (China)
Yihua Hu, Electronic Engineering Institute (China)
Xiaohong Tao, Electronic Engineering Institute (China)
Guilan Hu, Electronic Engineering Institute (China)

Published in SPIE Proceedings Vol. 5985:
International Conference on Space Information Technology
Cheng Wang; Shan Zhong; Xiulin Hu, Editor(s)

© SPIE. Terms of Use
Back to Top