Share Email Print
cover

Proceedings Paper

Utility of neural net classification for remote sensing data based on an improved image fusion algorithm
Author(s): Bofeng Cai; Rong Yu; Zengxiang Zhang
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

There are many different advantages and disadvantages in traditional subpixel classification methods such as uncertain classification accuracy, etc. which bring limitations for commonly application. In recent years, many algorithms have been used to resolve these problems. In this paper, based on an optimized image fusion algorithm, a comparison experiment on traditional maximum likelihood classification and neural net classification is performed. According to the classification accuracy data, the overall accuracy of classification increased from 81.67% to 89.67%.

Paper Details

Date Published: 28 October 2006
PDF: 10 pages
Proc. SPIE 6418, Geoinformatics 2006: GNSS and Integrated Geospatial Applications, 64180C (28 October 2006); doi: 10.1117/12.712584
Show Author Affiliations
Bofeng Cai, Institute of Remote Sensing Applications (China)
Rong Yu, Institute of Remote Sensing Applications (China)
Zengxiang Zhang, Institute of Remote Sensing Applications (China)


Published in SPIE Proceedings Vol. 6418:
Geoinformatics 2006: GNSS and Integrated Geospatial Applications
Deren Li; Linyuan Xia, Editor(s)

© SPIE. Terms of Use
Back to Top
PREMIUM CONTENT
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?
close_icon_gray