Please use this identifier to cite or link to this item:
Title: Classification of satellite images for land-cover changes using an unsupervised neural network algorithm
Authors: Hadjimitsis, Diofantos G. 
Evangelou, I.
Lazakidou, A.
Keywords: Neural networks;Remote-sensing images;Self-organizing maps
Issue Date: 2005
Source: WSEAS transactions on signal processing, 2005, Volume 1, Issue 2, Pages 155-162
Abstract: This paper presents a method for classifying Landsat Satellite Images. This method is based on the Self-Organizing Map (SOM), which is an unsupervised artificial neural network algorithm. The occurrence of fires on the island of Skiathos in Greece is considered as an example for investigating the possible impact of fires on Land-Cover changes. Classification is performed on three Landsat-5 TM satellite images of areas of the island, acquired in 1988, 1999 and 2000. Land-cover changes and areas affected by the fires are identified after the classification has been performed on the images. In order to assess the proposed SOM Artificial Neural network algorithm, the same images have been classified using conventional classification methods and their results are compared with the ones of the SOM.
ISSN: 1790-5052
Type: Article
Appears in Collections:Άρθρα/Articles

Show full item record

Page view(s) 50

Last Week
Last month
checked on Aug 18, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.