Examining region-based methods for land cover classification using stochastic distances
Autor(a) principal: | |
---|---|
Data de Publicação: | 2016 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1080/01431161.2016.1165883 http://hdl.handle.net/11449/168580 |
Resumo: | ABSTRACT: A recent alternative to standard pixel-based classification of remote-sensing data is region-based classification, which has proved to be particularly useful when analysing high-resolution imagery of complex environments, such as urban areas, or when addressing noisy data, such as synthetic aperture radar (SAR) images. First, following certain criteria, the imagery is decomposed into homogeneous regions, and then each region is classified into a class of interest. The usual method for region-based classification involves using stochastic distances, which measure the distances between the pixel distributions inside an unknown region and the representative distributions of each class. The class, which is at the minimum distance from the unknown region distribution, is assigned to the region and this procedure is termed stochastic minimum distance classification (SMDC). This study reports the use of methods derived from the original SMDC, Support Vector Machine (SVM), and graph theory, with the objective of identifying the most robust and accurate classification methods. The equivalent pixel-based versions of region-based analysed methods were included for comparison. A case study near the Tapajós National Forest, in Pará state, Brazil, was investigated using ALOS PALSAR data. This study showed that methods based on the nearest neighbour, derived from SMDC, and SVM, with a specific kernel function, are more accurate and robust than the other analysed methods for region-based classification. Furthermore, pixel-based methods are not indicated to perform the classification of images with a strong presence of noise, such as SAR images. |
id |
UNSP_37a55f4d66b59562ec0dd692201aeae6 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/168580 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Examining region-based methods for land cover classification using stochastic distancesABSTRACT: A recent alternative to standard pixel-based classification of remote-sensing data is region-based classification, which has proved to be particularly useful when analysing high-resolution imagery of complex environments, such as urban areas, or when addressing noisy data, such as synthetic aperture radar (SAR) images. First, following certain criteria, the imagery is decomposed into homogeneous regions, and then each region is classified into a class of interest. The usual method for region-based classification involves using stochastic distances, which measure the distances between the pixel distributions inside an unknown region and the representative distributions of each class. The class, which is at the minimum distance from the unknown region distribution, is assigned to the region and this procedure is termed stochastic minimum distance classification (SMDC). This study reports the use of methods derived from the original SMDC, Support Vector Machine (SVM), and graph theory, with the objective of identifying the most robust and accurate classification methods. The equivalent pixel-based versions of region-based analysed methods were included for comparison. A case study near the Tapajós National Forest, in Pará state, Brazil, was investigated using ALOS PALSAR data. This study showed that methods based on the nearest neighbour, derived from SMDC, and SVM, with a specific kernel function, are more accurate and robust than the other analysed methods for region-based classification. Furthermore, pixel-based methods are not indicated to perform the classification of images with a strong presence of noise, such as SAR images.Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Instituto de Ciência e Tecnologia UNESP – Univ. Estadual PaulistaDivisão de Processamento de Imagens INPE – Inst. Nacional de Pesquisas EspaciaisCenter for Global Change and Earth Observations MSU – Michigan State UniversityInstituto de Ciência e Tecnologia UNESP – Univ. Estadual PaulistaCNPq: 151571/2013-9FAPESP: 2014/14830-8CNPq: 307666/2011-5CNPq: 401528/2012-0Universidade Estadual Paulista (Unesp)INPE – Inst. Nacional de Pesquisas EspaciaisMSU – Michigan State UniversityNegri, R. G. [UNESP]Dutra, L. V.Sant'Anna, S. J.S.Lu, D.2018-12-11T16:42:01Z2018-12-11T16:42:01Z2016-04-17info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article1902-1921application/pdfhttp://dx.doi.org/10.1080/01431161.2016.1165883International Journal of Remote Sensing, v. 37, n. 8, p. 1902-1921, 2016.1366-59010143-1161http://hdl.handle.net/11449/16858010.1080/01431161.2016.11658832-s2.0-849638188812-s2.0-84963818881.pdfScopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengInternational Journal of Remote Sensing0,7960,796info:eu-repo/semantics/openAccess2023-11-07T06:09:10Zoai:repositorio.unesp.br:11449/168580Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462023-11-07T06:09:10Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Examining region-based methods for land cover classification using stochastic distances |
title |
Examining region-based methods for land cover classification using stochastic distances |
spellingShingle |
Examining region-based methods for land cover classification using stochastic distances Negri, R. G. [UNESP] |
title_short |
Examining region-based methods for land cover classification using stochastic distances |
title_full |
Examining region-based methods for land cover classification using stochastic distances |
title_fullStr |
Examining region-based methods for land cover classification using stochastic distances |
title_full_unstemmed |
Examining region-based methods for land cover classification using stochastic distances |
title_sort |
Examining region-based methods for land cover classification using stochastic distances |
author |
Negri, R. G. [UNESP] |
author_facet |
Negri, R. G. [UNESP] Dutra, L. V. Sant'Anna, S. J.S. Lu, D. |
author_role |
author |
author2 |
Dutra, L. V. Sant'Anna, S. J.S. Lu, D. |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) INPE – Inst. Nacional de Pesquisas Espaciais MSU – Michigan State University |
dc.contributor.author.fl_str_mv |
Negri, R. G. [UNESP] Dutra, L. V. Sant'Anna, S. J.S. Lu, D. |
description |
ABSTRACT: A recent alternative to standard pixel-based classification of remote-sensing data is region-based classification, which has proved to be particularly useful when analysing high-resolution imagery of complex environments, such as urban areas, or when addressing noisy data, such as synthetic aperture radar (SAR) images. First, following certain criteria, the imagery is decomposed into homogeneous regions, and then each region is classified into a class of interest. The usual method for region-based classification involves using stochastic distances, which measure the distances between the pixel distributions inside an unknown region and the representative distributions of each class. The class, which is at the minimum distance from the unknown region distribution, is assigned to the region and this procedure is termed stochastic minimum distance classification (SMDC). This study reports the use of methods derived from the original SMDC, Support Vector Machine (SVM), and graph theory, with the objective of identifying the most robust and accurate classification methods. The equivalent pixel-based versions of region-based analysed methods were included for comparison. A case study near the Tapajós National Forest, in Pará state, Brazil, was investigated using ALOS PALSAR data. This study showed that methods based on the nearest neighbour, derived from SMDC, and SVM, with a specific kernel function, are more accurate and robust than the other analysed methods for region-based classification. Furthermore, pixel-based methods are not indicated to perform the classification of images with a strong presence of noise, such as SAR images. |
publishDate |
2016 |
dc.date.none.fl_str_mv |
2016-04-17 2018-12-11T16:42:01Z 2018-12-11T16:42:01Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1080/01431161.2016.1165883 International Journal of Remote Sensing, v. 37, n. 8, p. 1902-1921, 2016. 1366-5901 0143-1161 http://hdl.handle.net/11449/168580 10.1080/01431161.2016.1165883 2-s2.0-84963818881 2-s2.0-84963818881.pdf |
url |
http://dx.doi.org/10.1080/01431161.2016.1165883 http://hdl.handle.net/11449/168580 |
identifier_str_mv |
International Journal of Remote Sensing, v. 37, n. 8, p. 1902-1921, 2016. 1366-5901 0143-1161 10.1080/01431161.2016.1165883 2-s2.0-84963818881 2-s2.0-84963818881.pdf |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
International Journal of Remote Sensing 0,796 0,796 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
1902-1921 application/pdf |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1797789633793228800 |