Share Email Print

Proceedings Paper

Neural networks for vegetation stress assessment
Author(s): Cristina Aiftimiei; Alexandra Caramizoiu; Axente D. Stoica; Aurelian Aiftimiei
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

In digital imagery, picture elements which contain two or more land cover classes are called mixed pixels. Small strips or patches of woody vegetation, typical landscape features in many farming areas, are frequently not detected by standard computer-assisted classification of digital imagery because such landscape features are smaller than the pixel size of the image and are mixed with other classes. Development of advanced techniques for improving remote sensing image classification accuracy is essential for deriving reliable land cover information for natural resource applications. Artificial neural networks ar among the optical tools for this type of application. Neural network techniques were developed for processing two dates of Landsat TM data, and measure of image texture for the purpose of deriving more accurate land use and land cover information. Classification results derive from the neural network approach overall were nearly 10 percent more accurate than those derived previously using a conventional maximum likelihood approach.

Paper Details

Date Published: 23 February 2000
PDF: 6 pages
Proc. SPIE 4068, SIOEL '99: Sixth Symposium on Optoelectronics, (23 February 2000); doi: 10.1117/12.378647
Show Author Affiliations
Cristina Aiftimiei, Institute of Optoelectronics (Romania)
Alexandra Caramizoiu, Institute of Optoelectronics (Romania)
Axente D. Stoica, Institute of Optoelectronics (Romania)
Aurelian Aiftimiei, Ministry of Defense (Romania)

Published in SPIE Proceedings Vol. 4068:
SIOEL '99: Sixth Symposium on Optoelectronics
Teodor Necsoiu; Maria Robu; Dan C. Dumitras, Editor(s)

© SPIE. Terms of Use
Back to Top