This study aims to determine whether artificial neural network classifiers are useful to extract land cover information from remotely sensed imagery in tropical savannas and how they perform compared with statistical procedures. The extent to which a neural network model could be used in sub-pixel classification and the use of environmental information related to the spatial distribution of the land cover categories of interest are investigated.
forestry, image processing, remote sensing
|De Wulf, Robert||promotor|
|Laboratory of Forest Management and Spatial Information Techniques||unknown|
created:2011-12-14 14:18:59 UTC, source:iweto