Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Neural networks, multitemporal landsat thematic mapper data and topographic data to classify forest damages in the Czech republic

Ardö, J. LU orcid ; Pilesjö, P. LU and Skidmore, A. LU (1997) In Canadian Journal of Remote Sensing 23(3). p.217-229
Abstract

This study uses multitemporal Landsat Thematic Mapper data and topographic data for the purpose of classifying coniferous forest damage in the Czech Republic using an artificial neural network. Comparing the neural network-based classification with earlier studies and a multinominal logistic regression using identical training and test data indicates that the back propagation algorithm is comparable, but not superior, to conventional methods. The dependence on the randomly set input weights and the more time-consuming back propagation training make neural network less useful for classification of forest damages than conventional classification algorithms. However, the ability to integrate and extract information from multisource data... (More)

This study uses multitemporal Landsat Thematic Mapper data and topographic data for the purpose of classifying coniferous forest damage in the Czech Republic using an artificial neural network. Comparing the neural network-based classification with earlier studies and a multinominal logistic regression using identical training and test data indicates that the back propagation algorithm is comparable, but not superior, to conventional methods. The dependence on the randomly set input weights and the more time-consuming back propagation training make neural network less useful for classification of forest damages than conventional classification algorithms. However, the ability to integrate and extract information from multisource data with different or unknown distributions are advantages of neural networks.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Canadian Journal of Remote Sensing
volume
23
issue
3
pages
13 pages
publisher
Taylor & Francis
external identifiers
  • scopus:0031221032
ISSN
0703-8992
DOI
10.1080/07038992.1997.10855204
language
English
LU publication?
yes
id
ae03ad86-fdd8-4264-b074-7d7592daddcf
date added to LUP
2016-04-07 14:27:22
date last changed
2022-01-30 02:28:17
@article{ae03ad86-fdd8-4264-b074-7d7592daddcf,
  abstract     = {{<p>This study uses multitemporal Landsat Thematic Mapper data and topographic data for the purpose of classifying coniferous forest damage in the Czech Republic using an artificial neural network. Comparing the neural network-based classification with earlier studies and a multinominal logistic regression using identical training and test data indicates that the back propagation algorithm is comparable, but not superior, to conventional methods. The dependence on the randomly set input weights and the more time-consuming back propagation training make neural network less useful for classification of forest damages than conventional classification algorithms. However, the ability to integrate and extract information from multisource data with different or unknown distributions are advantages of neural networks.</p>}},
  author       = {{Ardö, J. and Pilesjö, P. and Skidmore, A.}},
  issn         = {{0703-8992}},
  language     = {{eng}},
  number       = {{3}},
  pages        = {{217--229}},
  publisher    = {{Taylor & Francis}},
  series       = {{Canadian Journal of Remote Sensing}},
  title        = {{Neural networks, multitemporal landsat thematic mapper data and topographic data to classify forest damages in the Czech republic}},
  url          = {{http://dx.doi.org/10.1080/07038992.1997.10855204}},
  doi          = {{10.1080/07038992.1997.10855204}},
  volume       = {{23}},
  year         = {{1997}},
}