Paper Info
Title
Neural maps in remote sensing image analysis.
Abstract
We study the application of self-organizing maps (SOMs) for the analyses of remote sensing spectral images. Advanced airborne and satellite-based imaging spectrometers produce very high-dimensional spectral signatures that provide key information to many scientific investigations about the surface and atmosphere of Earth and other planets. These new, sophisticated data demand new and advanced approaches to cluster detection, visualization, and supervised classification. In this article we concentrate on the issue of faithful topological mapping in order to avoid false interpretations of cluster maps created by an SOM. We describe several new extensions of the standard SOM, developed in the past few years: the growing SOM, magnification control, and generalized relevance learning vector quantization, and demonstrate their effect on both low-dimensional traditional multi-spectral imagery and approximately 200-dimensional hyperspectral imagery.
Year
DOI
Venue
2003
10.1016/S0893-6080(03)00021-2
Neural Networks
Keywords
Field
DocType
remote sensing,cluster detection,200-dimensional hyperspectral imagery,faithful topological mapping,neural map,self-organizing map,standard som,cluster map,image analysis,high-dimensional spectral signature,low-dimensional traditional multi-spectral imagery,advanced approach,generalized relevance learning vector quantization,spectral image,new extension,imaging spectrometer,hyperspectral imagery,spectral imaging,self organizing map
Object detection,Computer vision,Visualization,Remote sensing,Hyperspectral imaging,Self-organizing map,Vector quantization,Artificial intelligence,Topological mapping,Artificial neural network,Spectral signature,Mathematics
Journal
Volume
Issue
ISSN
16
3-4
0893-6080
Citations 
PageRank 
References 
72
3.47
30
Authors
3
Name
Order
Citations
PageRank
Thomas Villmann11279118.19
Erzsébet Merényi218015.63
Barbara Hammer32383181.34