Title :
Sub-pixel land-cover classification with SPOT-VEGETATION imagery
Author :
Swinnen, Else ; Eerens, Herman ; Lissens, Gil ; Canters, Frank
Author_Institution :
Centre of Expertise on Remote Sensing & Atmos. Processes, Flemish Inst. for Technol. Res., Mol, Belgium
Abstract :
Knowledge about global land cover is an important input for the modelling of ecological and environmental processes. Production of such global vegetation maps can be facilitated by using automated methods for classification. Two neural network strategies, an overall and class-specific network(s), were tested on a part of Europe. This study indicates that sub-pixel proportion estimates can be assessed quite accurately from 1-km resolution SPOT-VEGETATION imagery
Keywords :
geophysical signal processing; image classification; neural nets; vegetation mapping; Europe; SPOT-VEGETATION imagery; automated methods; class-specific neural network; classification; global land cover; global vegetation maps; neural network strategies; overall neural network; sub-pixel land-cover classification; Biological system modeling; Continents; Gas insulated transmission lines; Image resolution; Neural networks; Production; Remote sensing; Spatial databases; Spatial resolution; Vegetation mapping;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2001. IGARSS '01. IEEE 2001 International
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-7031-7
DOI :
10.1109/IGARSS.2001.976214
