A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Tipo de documento: | Tese |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFPE |
Texto Completo: | https://repositorio.ufpe.br/handle/123456789/46587 |
Resumo: | Oil spills are among the most undesirable events in coastal environments because they are substantially harmful, with negative environmental, social, and economic consequences. In general, a risk framework for the event involves prevention, monitoring, detection, and damage mitigation. Regarding detection, rapid oil spill identification is essential for problem mitigation, which generally fosters the use of automated procedures. Usually, automated oil spill detection involves radar images, computer vision, and machine learning techniques to classify these images. In this work, we propose a novel image feature extraction method based on the q-Exponential probability distribution, named q-EFE. Such a probabilistic model is suitable to account for atypical extreme values of the variable of interest, e.g., pixels values, as it can have the power-law behavior. The q-EFE part is combined with machine learning methods to comprise a computer vision methodology to automatically classify images as “with oil spill” or “without oil spill”. Hence, we also propose a new automatic oil spill detection methodology that uses the q-EFE to rapidly identify oil spills in radar images. We used a public dataset composed of 1112 Synthetic Aperture Radar (SAR) images to validate our proposed methodology. Considering the proposed q-Exponential-based feature extraction, the tested Machine Learning methods and Deep Learning models architectures, the Support Vector Machine (SVM) and Extreme Gradient Boosting (XGB) models outperformed deep learning models and Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) techniques for the biggest dataset size. |
id |
UFPE_9d54d6bf6d99608d11a500aa51e2d6e0 |
---|---|
oai_identifier_str |
oai:repositorio.ufpe.br:123456789/46587 |
network_acronym_str |
UFPE |
network_name_str |
Repositório Institucional da UFPE |
repository_id_str |
2221 |
spelling |
NEGREIROS, Ana Cláudia Souza Vidal dehttp://lattes.cnpq.br/3480755550348791http://lattes.cnpq.br/5632602851077460LINS, Isis Didier2022-09-20T15:59:31Z2022-09-20T15:59:31Z2022-07-15NEGREIROS, Ana Cláudia Souza Vidal de. A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models. 2022. Tese (Doutorado em Engenharia de Produção) – Universidade Federal de Pernambuco, Recife, 2022.https://repositorio.ufpe.br/handle/123456789/46587Oil spills are among the most undesirable events in coastal environments because they are substantially harmful, with negative environmental, social, and economic consequences. In general, a risk framework for the event involves prevention, monitoring, detection, and damage mitigation. Regarding detection, rapid oil spill identification is essential for problem mitigation, which generally fosters the use of automated procedures. Usually, automated oil spill detection involves radar images, computer vision, and machine learning techniques to classify these images. In this work, we propose a novel image feature extraction method based on the q-Exponential probability distribution, named q-EFE. Such a probabilistic model is suitable to account for atypical extreme values of the variable of interest, e.g., pixels values, as it can have the power-law behavior. The q-EFE part is combined with machine learning methods to comprise a computer vision methodology to automatically classify images as “with oil spill” or “without oil spill”. Hence, we also propose a new automatic oil spill detection methodology that uses the q-EFE to rapidly identify oil spills in radar images. We used a public dataset composed of 1112 Synthetic Aperture Radar (SAR) images to validate our proposed methodology. Considering the proposed q-Exponential-based feature extraction, the tested Machine Learning methods and Deep Learning models architectures, the Support Vector Machine (SVM) and Extreme Gradient Boosting (XGB) models outperformed deep learning models and Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) techniques for the biggest dataset size.CAPESVazamentos de óleo estão entre os mais indesejáveis eventos que podem ocorrer em ambientes costeiros por causa de seu perigo substancial, com consequências ambientais, sociais e econômicas. Em geral ua metodologia de risco envolve prevenção, monitoramento, detecção e mitigação dos danos. A respeito da detecção, a rápida identificação de um vazamento de óleo é essencial para a mitigação dos problemas, que geralmente fomenta o uso de procedimentos automáticos. Usualmente, a detecção automática de vazamento de óleo envolve imagens de radar, visão computacional, e técnicas de aprendizado de máquina para classificar as imagens. Neste trabalho, um novo método de extração de características em imagens baseado na distribuição probabilística q-Exponencial, chamado de q-EFE, está sendo proposto. Esse modelo probabilístico é adequado para explicar valores extremos atípicos de variáveis de interesse, e.g., valores de pixels, uma vez que ele tem comportamento de lei de potência (power-law). A parte do q-EFE é combinada com métodos de aprendizado de máquina para compreender uma metodologia de visão computacional para classificar automaticamente imagens como “com vazamento de óleo” ou “sem vazamento de óleo”. Consequentemente, este trabalho propõe uma nova metodologia de detecção automática de vazamento de óleo que usa o q-EFE para identificar rapidamente vazamentos de óleo em imagens de radar. Foi utilizado um conjunto de dados composto por 1112 imagens geradas pelo Synthethic Aperture Radar (SAR) para validar a metodologia proposta. Considerando a extração de características proposta que é baseada na distribuição q-exponencial, os métodos de aprendizado de máquina e as arquiteturas dos modelos de aprendizado profundo testados, os modelos Support Vector Machine e Extreme Gradient Boosting (XGB) superaram os modelos deep learning e as técnicas de Local Binary Pattern (LBP) e Gray Level Co-occurrence Matrix (GLCM) para os maiores tamanhos de conjunto de dados.engUniversidade Federal de PernambucoPrograma de Pos Graduacao em Engenharia de ProducaoUFPEBrasilhttp://creativecommons.org/licenses/by-nc-nd/3.0/br/info:eu-repo/semantics/openAccessEngenharia de ProduçãoDistribuição q-ExponencialExtração de característicasAprendizado de máquinaVisão computacionalVazamentos de óleoAnálise de riscoA novel automated oil spill detection approach based on the q-Exponential distribution and machine learning modelsinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisdoutoradoreponame:Repositório Institucional da UFPEinstname:Universidade Federal de Pernambuco (UFPE)instacron:UFPETEXTTESE Ana Cláudia Souza Vidal de Negreiros.pdf.txtTESE Ana Cláudia Souza Vidal de Negreiros.pdf.txtExtracted texttext/plain251488https://repositorio.ufpe.br/bitstream/123456789/46587/4/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdf.txtd31ef391f124542f3273d3fc2283ba1fMD54THUMBNAILTESE Ana Cláudia Souza Vidal de Negreiros.pdf.jpgTESE Ana Cláudia Souza Vidal de Negreiros.pdf.jpgGenerated Thumbnailimage/jpeg1254https://repositorio.ufpe.br/bitstream/123456789/46587/5/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdf.jpgf05808452309939b48cf467134cd8e56MD55CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufpe.br/bitstream/123456789/46587/2/license_rdfe39d27027a6cc9cb039ad269a5db8e34MD52ORIGINALTESE Ana Cláudia Souza Vidal de Negreiros.pdfTESE Ana Cláudia Souza Vidal de Negreiros.pdfapplication/pdf7998896https://repositorio.ufpe.br/bitstream/123456789/46587/1/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdfd9f62c0a2ca80d6df3f10b7ba4bfe045MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-82142https://repositorio.ufpe.br/bitstream/123456789/46587/3/license.txt6928b9260b07fb2755249a5ca9903395MD53123456789/465872022-09-22 03:17:43.021oai:repositorio.ufpe.br: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ório InstitucionalPUBhttps://repositorio.ufpe.br/oai/requestattena@ufpe.bropendoar:22212022-09-22T06:17:43Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE)false |
dc.title.pt_BR.fl_str_mv |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
title |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
spellingShingle |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models NEGREIROS, Ana Cláudia Souza Vidal de Engenharia de Produção Distribuição q-Exponencial Extração de características Aprendizado de máquina Visão computacional Vazamentos de óleo Análise de risco |
title_short |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
title_full |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
title_fullStr |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
title_full_unstemmed |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
title_sort |
A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models |
author |
NEGREIROS, Ana Cláudia Souza Vidal de |
author_facet |
NEGREIROS, Ana Cláudia Souza Vidal de |
author_role |
author |
dc.contributor.authorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/3480755550348791 |
dc.contributor.advisorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/5632602851077460 |
dc.contributor.author.fl_str_mv |
NEGREIROS, Ana Cláudia Souza Vidal de |
dc.contributor.advisor1.fl_str_mv |
LINS, Isis Didier |
contributor_str_mv |
LINS, Isis Didier |
dc.subject.por.fl_str_mv |
Engenharia de Produção Distribuição q-Exponencial Extração de características Aprendizado de máquina Visão computacional Vazamentos de óleo Análise de risco |
topic |
Engenharia de Produção Distribuição q-Exponencial Extração de características Aprendizado de máquina Visão computacional Vazamentos de óleo Análise de risco |
description |
Oil spills are among the most undesirable events in coastal environments because they are substantially harmful, with negative environmental, social, and economic consequences. In general, a risk framework for the event involves prevention, monitoring, detection, and damage mitigation. Regarding detection, rapid oil spill identification is essential for problem mitigation, which generally fosters the use of automated procedures. Usually, automated oil spill detection involves radar images, computer vision, and machine learning techniques to classify these images. In this work, we propose a novel image feature extraction method based on the q-Exponential probability distribution, named q-EFE. Such a probabilistic model is suitable to account for atypical extreme values of the variable of interest, e.g., pixels values, as it can have the power-law behavior. The q-EFE part is combined with machine learning methods to comprise a computer vision methodology to automatically classify images as “with oil spill” or “without oil spill”. Hence, we also propose a new automatic oil spill detection methodology that uses the q-EFE to rapidly identify oil spills in radar images. We used a public dataset composed of 1112 Synthetic Aperture Radar (SAR) images to validate our proposed methodology. Considering the proposed q-Exponential-based feature extraction, the tested Machine Learning methods and Deep Learning models architectures, the Support Vector Machine (SVM) and Extreme Gradient Boosting (XGB) models outperformed deep learning models and Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) techniques for the biggest dataset size. |
publishDate |
2022 |
dc.date.accessioned.fl_str_mv |
2022-09-20T15:59:31Z |
dc.date.available.fl_str_mv |
2022-09-20T15:59:31Z |
dc.date.issued.fl_str_mv |
2022-07-15 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/doctoralThesis |
format |
doctoralThesis |
status_str |
publishedVersion |
dc.identifier.citation.fl_str_mv |
NEGREIROS, Ana Cláudia Souza Vidal de. A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models. 2022. Tese (Doutorado em Engenharia de Produção) – Universidade Federal de Pernambuco, Recife, 2022. |
dc.identifier.uri.fl_str_mv |
https://repositorio.ufpe.br/handle/123456789/46587 |
identifier_str_mv |
NEGREIROS, Ana Cláudia Souza Vidal de. A novel automated oil spill detection approach based on the q-Exponential distribution and machine learning models. 2022. Tese (Doutorado em Engenharia de Produção) – Universidade Federal de Pernambuco, Recife, 2022. |
url |
https://repositorio.ufpe.br/handle/123456789/46587 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
http://creativecommons.org/licenses/by-nc-nd/3.0/br/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nc-nd/3.0/br/ |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.publisher.program.fl_str_mv |
Programa de Pos Graduacao em Engenharia de Producao |
dc.publisher.initials.fl_str_mv |
UFPE |
dc.publisher.country.fl_str_mv |
Brasil |
publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFPE instname:Universidade Federal de Pernambuco (UFPE) instacron:UFPE |
instname_str |
Universidade Federal de Pernambuco (UFPE) |
instacron_str |
UFPE |
institution |
UFPE |
reponame_str |
Repositório Institucional da UFPE |
collection |
Repositório Institucional da UFPE |
bitstream.url.fl_str_mv |
https://repositorio.ufpe.br/bitstream/123456789/46587/4/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdf.txt https://repositorio.ufpe.br/bitstream/123456789/46587/5/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdf.jpg https://repositorio.ufpe.br/bitstream/123456789/46587/2/license_rdf https://repositorio.ufpe.br/bitstream/123456789/46587/1/TESE%20Ana%20Cl%c3%a1udia%20Souza%20Vidal%20de%20Negreiros.pdf https://repositorio.ufpe.br/bitstream/123456789/46587/3/license.txt |
bitstream.checksum.fl_str_mv |
d31ef391f124542f3273d3fc2283ba1f f05808452309939b48cf467134cd8e56 e39d27027a6cc9cb039ad269a5db8e34 d9f62c0a2ca80d6df3f10b7ba4bfe045 6928b9260b07fb2755249a5ca9903395 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE) |
repository.mail.fl_str_mv |
attena@ufpe.br |
_version_ |
1802310765691011072 |