Share Email Print

Proceedings Paper

Research on wetland classification approaches based on Hyperion hyperspectral image
Author(s): Xue Li; Weiguo Jiang
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

Land cover classification is essential for the monitoring, protecting and managing of wetlands. With Dongting Lake in Hunan province of China selected as the study area, a scene of hyperspectral image acquired by EO-1 Hyperion was used to evaluate the methods for land cover classification. After a series of preprocessing including bands removal, radiometric correction, strip removal and geometric registration, MNF transformation and band selection were adopted for dimension reducing. Then the image was classified into eight land cover types via MLC, SVM, SAM and MF classifier. Results reveal that higher classification accuracy would be obtained if data dimension reduction is done by MNF method. In addition, SVM performs best among the four classifiers, followed with MLC, while SAM and MF show worse performances.

Paper Details

Date Published: 30 October 2009
PDF: 8 pages
Proc. SPIE 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications, 74981O (30 October 2009); doi: 10.1117/12.833912
Show Author Affiliations
Xue Li, Beijing Normal Univ. (China)
Weiguo Jiang, Beijing Normal Univ. (China)

Published in SPIE Proceedings Vol. 7498:
MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications
Faxiong Zhang; Faxiong Zhang, Editor(s)

© SPIE. Terms of Use
Back to Top