File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Application of neural networks in forest ecological classification

TitleApplication of neural networks in forest ecological classification
Authors
Chen, JunGong, PengNie, J.Blais, J. A.R.
Issue Date1993
Citation
2013 ACSM/ASPRS Convention, New Orleans, LA, 1993. In Conference Proceedings, 1993, v. 3, p. 65-71 How to Cite?
AbstractForest ecosystem classification involves as its input data such information as forest and soil types which are at a nominal measurement scale. Conventional discriminant analysis algorithms which are only suitable for handling data at linear and ratio scales are not suitable for handling nominal scale data. An artificial neural netowrk with a back-propagation model, which can handle data at any measurement scale, has been applied to the forest ecosystem classification. Ground truth data and expert classification results have been used to train the neural network. Some preliminary results indicate that the neural network method has a greater discriminating power than a conventional minimum distance classifier. -Authors
Persistent Identifierhttp://hdl.handle.net/10722/296506

 

DC FieldValueLanguage
dc.contributor.authorChen, Jun-
dc.contributor.authorGong, Peng-
dc.contributor.authorNie, J.-
dc.contributor.authorBlais, J. A.R.-
dc.date.accessioned2021-02-25T15:16:03Z-
dc.date.available2021-02-25T15:16:03Z-
dc.date.issued1993-
dc.identifier.citation2013 ACSM/ASPRS Convention, New Orleans, LA, 1993. In Conference Proceedings, 1993, v. 3, p. 65-71-
dc.identifier.urihttp://hdl.handle.net/10722/296506-
dc.description.abstractForest ecosystem classification involves as its input data such information as forest and soil types which are at a nominal measurement scale. Conventional discriminant analysis algorithms which are only suitable for handling data at linear and ratio scales are not suitable for handling nominal scale data. An artificial neural netowrk with a back-propagation model, which can handle data at any measurement scale, has been applied to the forest ecosystem classification. Ground truth data and expert classification results have been used to train the neural network. Some preliminary results indicate that the neural network method has a greater discriminating power than a conventional minimum distance classifier. -Authors-
dc.languageeng-
dc.relation.ispartof2013 ACSM/ASPRS Convention-
dc.titleApplication of neural networks in forest ecological classification-
dc.typeConference_Paper-
dc.identifier.scopuseid_2-s2.0-0027846592-
dc.identifier.volume3-
dc.identifier.spage65-
dc.identifier.epage71-

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats