Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Outros Autores: | , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Brazilian Journal of Geology |
Texto Completo: | http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2317-48892021000400301 |
Resumo: | Abstract Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaóca (São Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m². The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution. |
id |
SBGEO-1_8df05f2c8b1245ca29fd2c47a1bf96e9 |
---|---|
oai_identifier_str |
oai:scielo:S2317-48892021000400301 |
network_acronym_str |
SBGEO-1 |
network_name_str |
Brazilian Journal of Geology |
repository_id_str |
|
spelling |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazilmass movementpixel classificationsupervised classificationRapidEye Multispectral imagedigital elevation modelAbstract Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaóca (São Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m². The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution.Sociedade Brasileira de Geologia2021-01-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersiontext/htmlhttp://old.scielo.br/scielo.php?script=sci_arttext&pid=S2317-48892021000400301Brazilian Journal of Geology v.51 n.4 2021reponame:Brazilian Journal of Geologyinstname:Sociedade Brasileira de Geologia (SBGEO)instacron:SBGEO10.1590/2317-4889202120200105info:eu-repo/semantics/openAccessDias,Helen CristinaSandre,Lucas HenriqueAlarcón,Diego Alejandro SatizábalGrohmann,Carlos HenriqueQuintanilha,José Albertoeng2021-12-14T00:00:00Zoai:scielo:S2317-48892021000400301Revistahttp://bjg.siteoficial.ws/index.htmhttps://old.scielo.br/oai/scielo-oai.phpsbgsede@sbgeo.org.br||claudio.riccomini@gmail.com2317-46922317-4692opendoar:2021-12-14T00:00Brazilian Journal of Geology - Sociedade Brasileira de Geologia (SBGEO)false |
dc.title.none.fl_str_mv |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
title |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
spellingShingle |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil Dias,Helen Cristina mass movement pixel classification supervised classification RapidEye Multispectral image digital elevation model |
title_short |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
title_full |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
title_fullStr |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
title_full_unstemmed |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
title_sort |
Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: A case study of Itaóca, southeastern Brazil |
author |
Dias,Helen Cristina |
author_facet |
Dias,Helen Cristina Sandre,Lucas Henrique Alarcón,Diego Alejandro Satizábal Grohmann,Carlos Henrique Quintanilha,José Alberto |
author_role |
author |
author2 |
Sandre,Lucas Henrique Alarcón,Diego Alejandro Satizábal Grohmann,Carlos Henrique Quintanilha,José Alberto |
author2_role |
author author author author |
dc.contributor.author.fl_str_mv |
Dias,Helen Cristina Sandre,Lucas Henrique Alarcón,Diego Alejandro Satizábal Grohmann,Carlos Henrique Quintanilha,José Alberto |
dc.subject.por.fl_str_mv |
mass movement pixel classification supervised classification RapidEye Multispectral image digital elevation model |
topic |
mass movement pixel classification supervised classification RapidEye Multispectral image digital elevation model |
description |
Abstract Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaóca (São Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m². The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021-01-01 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2317-48892021000400301 |
url |
http://old.scielo.br/scielo.php?script=sci_arttext&pid=S2317-48892021000400301 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
10.1590/2317-4889202120200105 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
text/html |
dc.publisher.none.fl_str_mv |
Sociedade Brasileira de Geologia |
publisher.none.fl_str_mv |
Sociedade Brasileira de Geologia |
dc.source.none.fl_str_mv |
Brazilian Journal of Geology v.51 n.4 2021 reponame:Brazilian Journal of Geology instname:Sociedade Brasileira de Geologia (SBGEO) instacron:SBGEO |
instname_str |
Sociedade Brasileira de Geologia (SBGEO) |
instacron_str |
SBGEO |
institution |
SBGEO |
reponame_str |
Brazilian Journal of Geology |
collection |
Brazilian Journal of Geology |
repository.name.fl_str_mv |
Brazilian Journal of Geology - Sociedade Brasileira de Geologia (SBGEO) |
repository.mail.fl_str_mv |
sbgsede@sbgeo.org.br||claudio.riccomini@gmail.com |
_version_ |
1752122399473532928 |