Sensoriamento Remoto aplicado em Recursos Hídricos
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Repositório Institucional da UFMS |
Texto Completo: | https://repositorio.ufms.br/handle/123456789/4656 |
Resumo: | The monitoring of water resources serves as a basis for decision making and even to mitigate the effects of future water crises, such as the crisis in the Cantareira System, the study area of this work, in the 2013/14 biennium. We investigated the reliability of image classification, using remote sensing techniques and machine learning in the context of water resources, which is an indispensable resource for society. The experiments were carried out in the six dams that make up the Cantareira System, and RapidEye orbital multispectral images were used, which have a spatial resolution of 5 meters. Four classification methods were tested, namely: Minimum Distance, Maximum Likelihood, Spectral Angle Mapping and Random Forest. The Minimum Distance and Maximum Likelihood methods offered results with an accuracy greater than 95%. The Random Forest, a machine learning technique, made it possible to generate results with superior accuracy, reaching an accuracy of 98.06%. The results show that the combination of RapidEye images with remote sensing and machine learning techniques allows detailed and accurate mapping of water resources in the Cantareira System. As a result of this research, there is also the generation of a set of labeled data, available for future experiments. |
id |
UFMS_9dc8f449a02431794bfc7d69a907653e |
---|---|
oai_identifier_str |
oai:repositorio.ufms.br:123456789/4656 |
network_acronym_str |
UFMS |
network_name_str |
Repositório Institucional da UFMS |
repository_id_str |
2124 |
spelling |
2022-04-06T13:23:26Z2022-04-06T13:23:26Z2021https://repositorio.ufms.br/handle/123456789/4656The monitoring of water resources serves as a basis for decision making and even to mitigate the effects of future water crises, such as the crisis in the Cantareira System, the study area of this work, in the 2013/14 biennium. We investigated the reliability of image classification, using remote sensing techniques and machine learning in the context of water resources, which is an indispensable resource for society. The experiments were carried out in the six dams that make up the Cantareira System, and RapidEye orbital multispectral images were used, which have a spatial resolution of 5 meters. Four classification methods were tested, namely: Minimum Distance, Maximum Likelihood, Spectral Angle Mapping and Random Forest. The Minimum Distance and Maximum Likelihood methods offered results with an accuracy greater than 95%. The Random Forest, a machine learning technique, made it possible to generate results with superior accuracy, reaching an accuracy of 98.06%. The results show that the combination of RapidEye images with remote sensing and machine learning techniques allows detailed and accurate mapping of water resources in the Cantareira System. As a result of this research, there is also the generation of a set of labeled data, available for future experiments.O monitoramento de recursos hídricos serve como base para tomada de decisão e para amenizar os efeitos de futuras crises hídricas, como, por exemplo, a crise no Sistema Cantareira, área de estudo deste trabalho, no biênio 2013/14. Investigamos a confiabilidade da classificação de imagens, utilizando técnicas de sensoriamento remoto e aprendizado de máquina no contexto de recursos hídricos, que é um recurso indispensável para a sociedade. Os experimentos foram realizados nas seis represas que compõem o Sistema Cantareira, e utilizamos imagens multiespectrais orbitais RapidEye, que apresentam uma resolução espacial de 5 metros. Foram testados quatro métodos de classificação, sendo eles: Distância Mínima, Probabilidade Máxima, Mapeamento do Ângulo Espectral e Random Forest. Os métodos de Distância Mínima e Probabilidade Máxima ofereceram resultados com acurácia maior que 95%. O Random Forest, técnica de aprendizado de máquina, possibilitou gerar resultados com acurácia superior, atingindo acurácia de 98,06%. Os resultados mostram que a combinação de imagens RapidEye com a as técnicas de sensoriamento remoto e aprendizado de máquina permitem o mapeamento detalhado e acurado de recursos hídricos no Sistema Cantareira. Como resultado dessa pesquisa, tem-se também a geração de um conjunto de dados rotulados, disponibilizado para a realização de experimentos futuros.Fundação Universidade Federal de Mato Grosso do SulUFMSBrasilSensoriamentoSensoriamento Remoto aplicado em Recursos Hídricosinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisJose Marcato JuniorLUCAS YURI DUTRA DE OLIVEIRAinfo:eu-repo/semantics/openAccessporreponame:Repositório Institucional da UFMSinstname:Universidade Federal de Mato Grosso do Sul (UFMS)instacron:UFMSTHUMBNAILSensoriamento Remoto e Aprendizado de Máquina aplicados em Recursos Hídricos.pdf.jpgSensoriamento Remoto e Aprendizado de Máquina aplicados em Recursos Hídricos.pdf.jpgGenerated Thumbnailimage/jpeg1268https://repositorio.ufms.br/bitstream/123456789/4656/3/Sensoriamento%20Remoto%20e%20Aprendizado%20de%20M%c3%a1quina%20aplicados%20em%20Recursos%20H%c3%addricos.pdf.jpg1932969bc1276fd4deddfc7d37267717MD53TEXTSensoriamento Remoto e Aprendizado de Máquina aplicados em Recursos Hídricos.pdf.txtSensoriamento Remoto e Aprendizado de Máquina aplicados em Recursos Hídricos.pdf.txtExtracted texttext/plain42761https://repositorio.ufms.br/bitstream/123456789/4656/2/Sensoriamento%20Remoto%20e%20Aprendizado%20de%20M%c3%a1quina%20aplicados%20em%20Recursos%20H%c3%addricos.pdf.txt9cb61681533ac64c3cf7ae5c66091299MD52123456789/46562022-04-12 08:51:19.078oai:repositorio.ufms.br:123456789/4656Repositório InstitucionalPUBhttps://repositorio.ufms.br/oai/requestri.prograd@ufms.bropendoar:21242022-04-12T12:51:19Repositório Institucional da UFMS - Universidade Federal de Mato Grosso do Sul (UFMS)false |
dc.title.pt_BR.fl_str_mv |
Sensoriamento Remoto aplicado em Recursos Hídricos |
title |
Sensoriamento Remoto aplicado em Recursos Hídricos |
spellingShingle |
Sensoriamento Remoto aplicado em Recursos Hídricos LUCAS YURI DUTRA DE OLIVEIRA Sensoriamento |
title_short |
Sensoriamento Remoto aplicado em Recursos Hídricos |
title_full |
Sensoriamento Remoto aplicado em Recursos Hídricos |
title_fullStr |
Sensoriamento Remoto aplicado em Recursos Hídricos |
title_full_unstemmed |
Sensoriamento Remoto aplicado em Recursos Hídricos |
title_sort |
Sensoriamento Remoto aplicado em Recursos Hídricos |
author |
LUCAS YURI DUTRA DE OLIVEIRA |
author_facet |
LUCAS YURI DUTRA DE OLIVEIRA |
author_role |
author |
dc.contributor.advisor1.fl_str_mv |
Jose Marcato Junior |
dc.contributor.author.fl_str_mv |
LUCAS YURI DUTRA DE OLIVEIRA |
contributor_str_mv |
Jose Marcato Junior |
dc.subject.por.fl_str_mv |
Sensoriamento |
topic |
Sensoriamento |
description |
The monitoring of water resources serves as a basis for decision making and even to mitigate the effects of future water crises, such as the crisis in the Cantareira System, the study area of this work, in the 2013/14 biennium. We investigated the reliability of image classification, using remote sensing techniques and machine learning in the context of water resources, which is an indispensable resource for society. The experiments were carried out in the six dams that make up the Cantareira System, and RapidEye orbital multispectral images were used, which have a spatial resolution of 5 meters. Four classification methods were tested, namely: Minimum Distance, Maximum Likelihood, Spectral Angle Mapping and Random Forest. The Minimum Distance and Maximum Likelihood methods offered results with an accuracy greater than 95%. The Random Forest, a machine learning technique, made it possible to generate results with superior accuracy, reaching an accuracy of 98.06%. The results show that the combination of RapidEye images with remote sensing and machine learning techniques allows detailed and accurate mapping of water resources in the Cantareira System. As a result of this research, there is also the generation of a set of labeled data, available for future experiments. |
publishDate |
2021 |
dc.date.issued.fl_str_mv |
2021 |
dc.date.accessioned.fl_str_mv |
2022-04-06T13:23:26Z |
dc.date.available.fl_str_mv |
2022-04-06T13:23:26Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://repositorio.ufms.br/handle/123456789/4656 |
url |
https://repositorio.ufms.br/handle/123456789/4656 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Fundação Universidade Federal de Mato Grosso do Sul |
dc.publisher.initials.fl_str_mv |
UFMS |
dc.publisher.country.fl_str_mv |
Brasil |
publisher.none.fl_str_mv |
Fundação Universidade Federal de Mato Grosso do Sul |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFMS instname:Universidade Federal de Mato Grosso do Sul (UFMS) instacron:UFMS |
instname_str |
Universidade Federal de Mato Grosso do Sul (UFMS) |
instacron_str |
UFMS |
institution |
UFMS |
reponame_str |
Repositório Institucional da UFMS |
collection |
Repositório Institucional da UFMS |
bitstream.url.fl_str_mv |
https://repositorio.ufms.br/bitstream/123456789/4656/3/Sensoriamento%20Remoto%20e%20Aprendizado%20de%20M%c3%a1quina%20aplicados%20em%20Recursos%20H%c3%addricos.pdf.jpg https://repositorio.ufms.br/bitstream/123456789/4656/2/Sensoriamento%20Remoto%20e%20Aprendizado%20de%20M%c3%a1quina%20aplicados%20em%20Recursos%20H%c3%addricos.pdf.txt |
bitstream.checksum.fl_str_mv |
1932969bc1276fd4deddfc7d37267717 9cb61681533ac64c3cf7ae5c66091299 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFMS - Universidade Federal de Mato Grosso do Sul (UFMS) |
repository.mail.fl_str_mv |
ri.prograd@ufms.br |
_version_ |
1807552836069752832 |