Change detection in forests and savannas using statistical analysis based on geographical objects

Detalhes bibliográficos
Autor(a) principal: Leite, Lucilia Rezende
Data de Publicação: 2017
Outros Autores: Carvalho, Luis Marcelo Tavares de, Silva, Fortunato Menezes da
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Institucional da UFLA
Texto Completo: http://repositorio.ufla.br/jspui/handle/1/30960
Resumo: The aim of this work was to assess techniques of land cover change detection in areas of Brazilian Forest and Savanna, using Landsat 5/TM images, and two iterative statistical methodologies based on geographical objects. The sensitivity of the methodologies was assessed in relation to the heterogeneity of the input data, the use of reflectance data and vegetation indices, and the use of different levels of confidence. The periods analyzed were from 2000 to 2006, and from 2006 to 2010. After the segmentation of images, the descriptive statistics average and standard deviation of each object were extracted. The determination of change objects was realized in an iterative way based on the Mahalanobis Distance and the chi-square distribution. The results were validated with an early visual detection and analyzed according to Receiver Operating Characteristic (ROC) Curve. Significant gains were obtained by using vegetation masks and bands 3 and 4 for both areas tested with 94,67% and 95,02% of the objects correctly detected as changes, respectively for the areas of Forest and Savanna. The use of the NDVI and different images were not satisfactory in this study.
id UFLA_e21fc8e8421aa754d9655cc5b004ae1a
oai_identifier_str oai:localhost:1/30960
network_acronym_str UFLA
network_name_str Repositório Institucional da UFLA
repository_id_str
spelling Change detection in forests and savannas using statistical analysis based on geographical objectsDetecção de mudanças em florestas e savanas utilizando análise estatística baseada em objetos geográficosBrazilian savannaAmazon forestRemote senseSegmentation of imagesDistance of MahalanobisSavana brasileiraFloresta AmazônicaSensoriamento remotoSegmentação de imagensDistância de MahalanobisThe aim of this work was to assess techniques of land cover change detection in areas of Brazilian Forest and Savanna, using Landsat 5/TM images, and two iterative statistical methodologies based on geographical objects. The sensitivity of the methodologies was assessed in relation to the heterogeneity of the input data, the use of reflectance data and vegetation indices, and the use of different levels of confidence. The periods analyzed were from 2000 to 2006, and from 2006 to 2010. After the segmentation of images, the descriptive statistics average and standard deviation of each object were extracted. The determination of change objects was realized in an iterative way based on the Mahalanobis Distance and the chi-square distribution. The results were validated with an early visual detection and analyzed according to Receiver Operating Characteristic (ROC) Curve. Significant gains were obtained by using vegetation masks and bands 3 and 4 for both areas tested with 94,67% and 95,02% of the objects correctly detected as changes, respectively for the areas of Forest and Savanna. The use of the NDVI and different images were not satisfactory in this study.O objetivo deste trabalho foi avaliar a detecção de mudanças sobre a cobertura do solo em áreas de Floresta e Savana Brasileira, utilizando imagens do satélite Landsat 5/TM e duas metodologias estatísticas iterativas baseadas em objetos geográficos. Foi avaliada a sensibilidade das metodologias em relação à heterogeneidade dos dados de entrada, à utilização de dados de reflectância e índices de vegetação e à utilização de diferentes níveis de confiança. Os períodos analisados compreenderam os anos 2000 a 2006 e 2006 a 2010. Após a segmentação das imagens foram extraídas as grandezas estatísticas descritivas média e desvio padrão de cada objeto. A determinação dos objetos de mudança foi realizada de forma iterativa com base na Distância de Mahalanobis e na distribuição qui-quadrado. Os resultados foram validados com uma prévia detecção visual e analisados de acordo com a curva ROC. Foram obtidos ganhos significativos na utilização de máscara e das bandas 3 e 4 para ambas as áreas testadas com 94,67% e 95,02% dos objetos corretamente detectados como mudança, respectivamente para as áreas de Floresta e Savana. O uso do NDVI e de imagens diferentes se mostraram insatisfatórios para a detecção de mudanças nas áreas testadas.Universidade Federal do Paraná2018-10-08T19:14:15Z2018-10-08T19:14:15Z2017info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfLEITE, L. R.; CARVALHO, L. M. T. de; SILVA, F. M. da. Change detection in forests and savannas using statistical analysis based on geographical objects. Boletim de Ciências Geodésicas, Curitiba, v. 23, n. 2, p. 284 - 295, Apr./June 2017.http://repositorio.ufla.br/jspui/handle/1/30960Boletim de Ciências Geodésicasreponame:Repositório Institucional da UFLAinstname:Universidade Federal de Lavras (UFLA)instacron:UFLAAttribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessLeite, Lucilia RezendeCarvalho, Luis Marcelo Tavares deSilva, Fortunato Menezes daeng2023-05-09T17:31:48Zoai:localhost:1/30960Repositório InstitucionalPUBhttp://repositorio.ufla.br/oai/requestnivaldo@ufla.br || repositorio.biblioteca@ufla.bropendoar:2023-05-09T17:31:48Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA)false
dc.title.none.fl_str_mv Change detection in forests and savannas using statistical analysis based on geographical objects
Detecção de mudanças em florestas e savanas utilizando análise estatística baseada em objetos geográficos
title Change detection in forests and savannas using statistical analysis based on geographical objects
spellingShingle Change detection in forests and savannas using statistical analysis based on geographical objects
Leite, Lucilia Rezende
Brazilian savanna
Amazon forest
Remote sense
Segmentation of images
Distance of Mahalanobis
Savana brasileira
Floresta Amazônica
Sensoriamento remoto
Segmentação de imagens
Distância de Mahalanobis
title_short Change detection in forests and savannas using statistical analysis based on geographical objects
title_full Change detection in forests and savannas using statistical analysis based on geographical objects
title_fullStr Change detection in forests and savannas using statistical analysis based on geographical objects
title_full_unstemmed Change detection in forests and savannas using statistical analysis based on geographical objects
title_sort Change detection in forests and savannas using statistical analysis based on geographical objects
author Leite, Lucilia Rezende
author_facet Leite, Lucilia Rezende
Carvalho, Luis Marcelo Tavares de
Silva, Fortunato Menezes da
author_role author
author2 Carvalho, Luis Marcelo Tavares de
Silva, Fortunato Menezes da
author2_role author
author
dc.contributor.author.fl_str_mv Leite, Lucilia Rezende
Carvalho, Luis Marcelo Tavares de
Silva, Fortunato Menezes da
dc.subject.por.fl_str_mv Brazilian savanna
Amazon forest
Remote sense
Segmentation of images
Distance of Mahalanobis
Savana brasileira
Floresta Amazônica
Sensoriamento remoto
Segmentação de imagens
Distância de Mahalanobis
topic Brazilian savanna
Amazon forest
Remote sense
Segmentation of images
Distance of Mahalanobis
Savana brasileira
Floresta Amazônica
Sensoriamento remoto
Segmentação de imagens
Distância de Mahalanobis
description The aim of this work was to assess techniques of land cover change detection in areas of Brazilian Forest and Savanna, using Landsat 5/TM images, and two iterative statistical methodologies based on geographical objects. The sensitivity of the methodologies was assessed in relation to the heterogeneity of the input data, the use of reflectance data and vegetation indices, and the use of different levels of confidence. The periods analyzed were from 2000 to 2006, and from 2006 to 2010. After the segmentation of images, the descriptive statistics average and standard deviation of each object were extracted. The determination of change objects was realized in an iterative way based on the Mahalanobis Distance and the chi-square distribution. The results were validated with an early visual detection and analyzed according to Receiver Operating Characteristic (ROC) Curve. Significant gains were obtained by using vegetation masks and bands 3 and 4 for both areas tested with 94,67% and 95,02% of the objects correctly detected as changes, respectively for the areas of Forest and Savanna. The use of the NDVI and different images were not satisfactory in this study.
publishDate 2017
dc.date.none.fl_str_mv 2017
2018-10-08T19:14:15Z
2018-10-08T19:14:15Z
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 LEITE, L. R.; CARVALHO, L. M. T. de; SILVA, F. M. da. Change detection in forests and savannas using statistical analysis based on geographical objects. Boletim de Ciências Geodésicas, Curitiba, v. 23, n. 2, p. 284 - 295, Apr./June 2017.
http://repositorio.ufla.br/jspui/handle/1/30960
identifier_str_mv LEITE, L. R.; CARVALHO, L. M. T. de; SILVA, F. M. da. Change detection in forests and savannas using statistical analysis based on geographical objects. Boletim de Ciências Geodésicas, Curitiba, v. 23, n. 2, p. 284 - 295, Apr./June 2017.
url http://repositorio.ufla.br/jspui/handle/1/30960
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade Federal do Paraná
publisher.none.fl_str_mv Universidade Federal do Paraná
dc.source.none.fl_str_mv Boletim de Ciências Geodésicas
reponame:Repositório Institucional da UFLA
instname:Universidade Federal de Lavras (UFLA)
instacron:UFLA
instname_str Universidade Federal de Lavras (UFLA)
instacron_str UFLA
institution UFLA
reponame_str Repositório Institucional da UFLA
collection Repositório Institucional da UFLA
repository.name.fl_str_mv Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA)
repository.mail.fl_str_mv nivaldo@ufla.br || repositorio.biblioteca@ufla.br
_version_ 1807835147426332672

Registros relacionados