Bitemporal analysis of burnt areas in the Atlantic Forest
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , |
Tipo de documento: | Artigo |
Idioma: | por eng |
Título da fonte: | Sociedade & natureza (Online) |
Texto Completo: | https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339 |
Resumo: | The study of burned areas is used as a subsidy for fire control and monitoring in the protected areas. In face of the challenges of the spectral signature characterization of burned areas, this study aimed to apply the object-oriented classification method and to evaluate the performance of spectral indices subsets for mapping burned areas in the Atlantic Forest. For that, we performed a bitemporal analysis between 2014 and 2016, considering the difference of each spectral indices among two LANDSAT 8 images: pre-and post-fire. The object-oriented classification was performed automatically by segmentation, supervised classification and optimization algorithms in the GIS environment. The “weak” burn severity class was the most expressive, with 13.65% of the mapped area, while the “severe” burn severity class occupied 0.3%. The burned areas presented an increase of reflectance in the red and shortwave infrared bands and a decrease in the near infrared band. The ΔNBR was the best discriminator of burned area and the ΔNBR, ΔNBR2, ΔNDMI, ΔSAVI, ΔNDVI, ΔGEMI and ΔMSAVI set presented the highest separation threshold. The validation of the classification by the Kappa agreement coefficient obtained a good outcome (0.72). The selection of the variables showed efficiency in determining the spectral indices’ subset with the best performance for detecting the classes of burned areas, improving the classification accuracy and reliability. The segmentation was also important for the effectiveness of the object-oriented classification, being directly influenced by the image spatial resolution. |
id |
UFU-6_2c3fd7f8793133bdbb9715a7d7406d44 |
---|---|
oai_identifier_str |
oai:ojs.www.seer.ufu.br:article/53339 |
network_acronym_str |
UFU-6 |
network_name_str |
Sociedade & natureza (Online) |
repository_id_str |
|
spelling |
Bitemporal analysis of burnt areas in the Atlantic ForestAnálise bitemporal de áreas queimadas na Mata AtlânticaClassificação de imagemSegmentação de imagemRazão de bandasSeleção de variáveisIncêndios florestais.Image classificationImage segmentationBand ratioVariables selectionForest fireThe study of burned areas is used as a subsidy for fire control and monitoring in the protected areas. In face of the challenges of the spectral signature characterization of burned areas, this study aimed to apply the object-oriented classification method and to evaluate the performance of spectral indices subsets for mapping burned areas in the Atlantic Forest. For that, we performed a bitemporal analysis between 2014 and 2016, considering the difference of each spectral indices among two LANDSAT 8 images: pre-and post-fire. The object-oriented classification was performed automatically by segmentation, supervised classification and optimization algorithms in the GIS environment. The “weak” burn severity class was the most expressive, with 13.65% of the mapped area, while the “severe” burn severity class occupied 0.3%. The burned areas presented an increase of reflectance in the red and shortwave infrared bands and a decrease in the near infrared band. The ΔNBR was the best discriminator of burned area and the ΔNBR, ΔNBR2, ΔNDMI, ΔSAVI, ΔNDVI, ΔGEMI and ΔMSAVI set presented the highest separation threshold. The validation of the classification by the Kappa agreement coefficient obtained a good outcome (0.72). The selection of the variables showed efficiency in determining the spectral indices’ subset with the best performance for detecting the classes of burned areas, improving the classification accuracy and reliability. The segmentation was also important for the effectiveness of the object-oriented classification, being directly influenced by the image spatial resolution.O estudo de áreas queimadas serve como subsídio para os planos de controle e monitoramento do fogo nas unidades de conservação. Diante dos desafios de caracterização do comportamento espectral de áreas queimadas, este estudo objetivou aplicar o método de classificação orientada a objeto e avaliar a melhor performance do uso conjunto de índices espectrais para o mapeamento de área queimadas na Mata Atlântica do Sul da Bahia. Para tanto, foi realizada análise bitemporal entre 2014 e 2016, considerando a diferença de cada índice espectral entre duas imagens LANDSAT 8: pré e pós-fogo. A classificação orientada a objeto foi executada de maneira supervisionada e automatizada por meio de algoritmos de segmentação, classificação e otimização em ambiente SIG. A classe de intensidade de queimada fraca foi a mais expressiva, com 13,65% da área mapeada, enquanto a classe severa ocupou apenas 0,3%. As áreas queimadas apresentaram um aumento da reflectância na faixa do vermelho e do infravermelho médio e uma diminuição na faixa do infravermelho próximo. O ΔNBR foi o melhor discriminador de áreas queimadas e o conjunto ΔNBR, ΔNBR2, ΔNDMI, ΔSAVI, ΔNDVI, ΔGEMI e ΔMSAVI, apresentou o maior limiar de separação. A validação da classificação feita pelo coeficiente de concordância Kappa obteve um bom resultado (0,72). A seleção de variáveis mostrou-se eficiente na determinação do conjunto de índices espectrais de melhor performance para detecção das classes de queimadas, melhorando a acurácia da classificação e a confiabilidade dos resultados. A etapa de segmentação também se mostrou importante para a eficácia da classificação orientada a objeto, sendo influenciada diretamente pela resolução espacial da imagem.Universidade Federal de Uberlândia2020-08-14info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://seer.ufu.br/index.php/sociedadenatureza/article/view/5333910.14393/SN-v32-2020-53339Sociedade & Natureza; Vol. 32 (2020); 565-577Sociedade & Natureza; v. 32 (2020); 565-5771982-45130103-1570reponame:Sociedade & natureza (Online)instname:Universidade Federal de Uberlândia (UFU)instacron:UFUporenghttps://seer.ufu.br/index.php/sociedadenatureza/article/view/53339/29647https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339/29648Copyright (c) 2020 Iorrana Figueiredo Sacramento, Roberto Ferreira Machado Michel, Rafael Gomes Siqueirainfo:eu-repo/semantics/openAccessSacramento, Iorrana FigueiredoMichel, Roberto Ferreira MachadoSiqueira, Rafael Gomes2021-07-01T16:18:20Zoai:ojs.www.seer.ufu.br:article/53339Revistahttp://www.sociedadenatureza.ig.ufu.br/PUBhttps://seer.ufu.br/index.php/sociedadenatureza/oai||sociedade.natureza.ufu@gmail.com|| lucianamelo@ufu.br1982-45130103-1570opendoar:2021-07-01T16:18:20Sociedade & natureza (Online) - Universidade Federal de Uberlândia (UFU)false |
dc.title.none.fl_str_mv |
Bitemporal analysis of burnt areas in the Atlantic Forest Análise bitemporal de áreas queimadas na Mata Atlântica |
title |
Bitemporal analysis of burnt areas in the Atlantic Forest |
spellingShingle |
Bitemporal analysis of burnt areas in the Atlantic Forest Sacramento, Iorrana Figueiredo Classificação de imagem Segmentação de imagem Razão de bandas Seleção de variáveis Incêndios florestais. Image classification Image segmentation Band ratio Variables selection Forest fire |
title_short |
Bitemporal analysis of burnt areas in the Atlantic Forest |
title_full |
Bitemporal analysis of burnt areas in the Atlantic Forest |
title_fullStr |
Bitemporal analysis of burnt areas in the Atlantic Forest |
title_full_unstemmed |
Bitemporal analysis of burnt areas in the Atlantic Forest |
title_sort |
Bitemporal analysis of burnt areas in the Atlantic Forest |
author |
Sacramento, Iorrana Figueiredo |
author_facet |
Sacramento, Iorrana Figueiredo Michel, Roberto Ferreira Machado Siqueira, Rafael Gomes |
author_role |
author |
author2 |
Michel, Roberto Ferreira Machado Siqueira, Rafael Gomes |
author2_role |
author author |
dc.contributor.author.fl_str_mv |
Sacramento, Iorrana Figueiredo Michel, Roberto Ferreira Machado Siqueira, Rafael Gomes |
dc.subject.por.fl_str_mv |
Classificação de imagem Segmentação de imagem Razão de bandas Seleção de variáveis Incêndios florestais. Image classification Image segmentation Band ratio Variables selection Forest fire |
topic |
Classificação de imagem Segmentação de imagem Razão de bandas Seleção de variáveis Incêndios florestais. Image classification Image segmentation Band ratio Variables selection Forest fire |
description |
The study of burned areas is used as a subsidy for fire control and monitoring in the protected areas. In face of the challenges of the spectral signature characterization of burned areas, this study aimed to apply the object-oriented classification method and to evaluate the performance of spectral indices subsets for mapping burned areas in the Atlantic Forest. For that, we performed a bitemporal analysis between 2014 and 2016, considering the difference of each spectral indices among two LANDSAT 8 images: pre-and post-fire. The object-oriented classification was performed automatically by segmentation, supervised classification and optimization algorithms in the GIS environment. The “weak” burn severity class was the most expressive, with 13.65% of the mapped area, while the “severe” burn severity class occupied 0.3%. The burned areas presented an increase of reflectance in the red and shortwave infrared bands and a decrease in the near infrared band. The ΔNBR was the best discriminator of burned area and the ΔNBR, ΔNBR2, ΔNDMI, ΔSAVI, ΔNDVI, ΔGEMI and ΔMSAVI set presented the highest separation threshold. The validation of the classification by the Kappa agreement coefficient obtained a good outcome (0.72). The selection of the variables showed efficiency in determining the spectral indices’ subset with the best performance for detecting the classes of burned areas, improving the classification accuracy and reliability. The segmentation was also important for the effectiveness of the object-oriented classification, being directly influenced by the image spatial resolution. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020-08-14 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339 10.14393/SN-v32-2020-53339 |
url |
https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339 |
identifier_str_mv |
10.14393/SN-v32-2020-53339 |
dc.language.iso.fl_str_mv |
por eng |
language |
por eng |
dc.relation.none.fl_str_mv |
https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339/29647 https://seer.ufu.br/index.php/sociedadenatureza/article/view/53339/29648 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Uberlândia |
publisher.none.fl_str_mv |
Universidade Federal de Uberlândia |
dc.source.none.fl_str_mv |
Sociedade & Natureza; Vol. 32 (2020); 565-577 Sociedade & Natureza; v. 32 (2020); 565-577 1982-4513 0103-1570 reponame:Sociedade & natureza (Online) instname:Universidade Federal de Uberlândia (UFU) instacron:UFU |
instname_str |
Universidade Federal de Uberlândia (UFU) |
instacron_str |
UFU |
institution |
UFU |
reponame_str |
Sociedade & natureza (Online) |
collection |
Sociedade & natureza (Online) |
repository.name.fl_str_mv |
Sociedade & natureza (Online) - Universidade Federal de Uberlândia (UFU) |
repository.mail.fl_str_mv |
||sociedade.natureza.ufu@gmail.com|| lucianamelo@ufu.br |
_version_ |
1799943981257195520 |