Meta-learning applications in digital image processing
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Biblioteca Digital de Teses e Dissertações da UFMA |
Texto Completo: | https://tedebc.ufma.br/jspui/handle/tede/tede/2934 |
Resumo: | In recent decades, advances in capture devices and increase of available digital image data have stimulated the creation of methodologies for data processing that produce various forms of valuable models, such as descriptors, classifiers, approximations and visualizations. These models are often developed in the field of machine learning, which is characterized by a large number of available algorithms, these algorithms often do not have guidelines to identify the most appropriate one based on specific data to which they will be applied and nature of problem under analysis. There is a knowledge that allows to relate the features of the algorithms and data that present a good performance to fulfill a specific task, known as Meta-Knowledge, which can include information on algorithms, evaluation metrics to calculate similarity of datasets or relation of tasks. Being Meta-Learning the study of methods based on principles that explore the Meta-Knowledge to obtain efficient models and solutions, adapting the processes of Machine Learning and Data Mining. The research carried out in this work analyzes the applications and advantages offered by Meta-Learning in field of digital image processing. To carry out this task, different types of images, characterizers, and feature analysis techniques are used; in addition, multiple Machine Learning techniques are applied. The results obtained show that methodology based on Meta-Learning is efficient when applied in processing of digital images for identification and storage of experience generated by developing methodologies for classification of different types of images, obtaining a high performance with respect to an evaluation metrics. This statement means that Meta-Learning allows recommending the most appropriate methodology to perform the processing of a specific type of image based on features of dataset under analysis and the type of specific task to be performed. |
id |
UFMA_32bb16054cc593e2057dfd5aaf3bc6fe |
---|---|
oai_identifier_str |
oai:tede2:tede/2934 |
network_acronym_str |
UFMA |
network_name_str |
Biblioteca Digital de Teses e Dissertações da UFMA |
repository_id_str |
2131 |
spelling |
SILVA, Aristófanes Corrêa288745363-72http://lattes.cnpq.br/2446301582459104SILVA, Aristófanes Corrêa288745363-72http://lattes.cnpq.br/2446301582459104CONCI, Aurahttps://orcid.org/0000-0003-0782-2501http://lattes.cnpq.br/5601388085745497LOPES, Denivaldo Cicero Pavaohttp://lattes.cnpq.br/7611180871627212ALMEIDA, João Dallyson Sousa dehttp://lattes.cnpq.br/6047330108382641629735713-75SEPULVEDA, Luis Fernando Marin2019-11-27T13:45:51Z2019-11-08SEPULVEDA, Luis Fernando Marin. Meta-learning applications in digital image processing. 2019. 112 f. Dissertação (Programa de Pós-Graduação em Engenharia de Eletricidade/CCET) - Universidade Federal do Maranhão, São Luís, 2019 .https://tedebc.ufma.br/jspui/handle/tede/tede/2934In recent decades, advances in capture devices and increase of available digital image data have stimulated the creation of methodologies for data processing that produce various forms of valuable models, such as descriptors, classifiers, approximations and visualizations. These models are often developed in the field of machine learning, which is characterized by a large number of available algorithms, these algorithms often do not have guidelines to identify the most appropriate one based on specific data to which they will be applied and nature of problem under analysis. There is a knowledge that allows to relate the features of the algorithms and data that present a good performance to fulfill a specific task, known as Meta-Knowledge, which can include information on algorithms, evaluation metrics to calculate similarity of datasets or relation of tasks. Being Meta-Learning the study of methods based on principles that explore the Meta-Knowledge to obtain efficient models and solutions, adapting the processes of Machine Learning and Data Mining. The research carried out in this work analyzes the applications and advantages offered by Meta-Learning in field of digital image processing. To carry out this task, different types of images, characterizers, and feature analysis techniques are used; in addition, multiple Machine Learning techniques are applied. The results obtained show that methodology based on Meta-Learning is efficient when applied in processing of digital images for identification and storage of experience generated by developing methodologies for classification of different types of images, obtaining a high performance with respect to an evaluation metrics. This statement means that Meta-Learning allows recommending the most appropriate methodology to perform the processing of a specific type of image based on features of dataset under analysis and the type of specific task to be performed.Nas últimas décadas, avanços nos dispositivos de captura e aumento da imagem digital disponível dados estimularam a criação de metodologias para processamento de dados que produzem várias formas de modelos valiosos, como descritores, classificadores, aproximações e visualizações. Esses modelos são freqüentemente desenvolvidos no campo de aprendizado de máquina, que é caracterizado por um grande número de algoritmos disponíveis, esses algoritmos geralmente não possui diretrizes para identificar a mais apropriada com base em dados específicos aos quais eles serão aplicados e a natureza do problema em análise. Existe um conhecimento que permite relacionar os recursos dos algoritmos e dados que apresentam um bom desempenho para realizar uma tarefa específica, conhecida como meta-conhecimento, que pode incluir informações sobre algoritmos, métricas de avaliação para calcular a similaridade de conjuntos de dados ou a relação de tarefas. Ser Meta-Aprender o estudo de métodos baseados em princípios que exploram o Meta-Conhecimento obter modelos e soluções eficientes, adaptando os processos de Machine Learning e mineração de dados. A pesquisa realizada neste trabalho analisa as aplicações e vantagens oferecidas pelo Meta-Learning no campo do processamento de imagens digitais. Para realizar Nesta tarefa, diferentes tipos de imagens, caracterizadores e técnicas de análise de recursos são usava; além disso, várias técnicas de aprendizado de máquina são aplicadas. Os resultados obtidos mostram que a metodologia baseada no Meta-Learning é eficiente quando aplicada no processamento de imagens digitais para identificação e armazenamento da experiência gerada pelo desenvolvimento metodologias para classificação de diferentes tipos de imagens, obtendo um alto desempenho com relação a uma métrica de avaliação. Esta afirmação significa que o Meta-Learning permite recomendar a metodologia mais apropriada para executar o processamento de um tipo de imagem com base nos recursos do conjunto de dados em análise e no tipo de tarefa específica a ser ser realizado.Submitted by Daniella Santos (daniella.santos@ufma.br) on 2019-11-27T13:45:51Z No. of bitstreams: 1 LuisFernandoSepulveda.pdf: 6098251 bytes, checksum: 4e29a071a0a57272848ee7858986fdc0 (MD5)Made available in DSpace on 2019-11-27T13:45:51Z (GMT). No. of bitstreams: 1 LuisFernandoSepulveda.pdf: 6098251 bytes, checksum: 4e29a071a0a57272848ee7858986fdc0 (MD5) Previous issue date: 2019-11-08application/pdfporUniversidade Federal do MaranhãoPROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE ELETRICIDADE/CCETUFMABrasilDEPARTAMENTO DE ENGENHARIA DA ELETRICIDADE/CCETMeta-learningImage processingMachine-learningFeature selectionMeta-dataCNNMeta-aprendizagemProcessamento de imagemAprendizado de máquinaSeleção de recursosMeta-dadosCNNEngenharia ElétricaMeta-learning applications in digital image processingAplicações de meta-aprendizado no processamento de imagens digitaisinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/openAccessreponame:Biblioteca Digital de Teses e Dissertações da UFMAinstname:Universidade Federal do Maranhão (UFMA)instacron:UFMAORIGINALLuisFernandoSepulveda.pdfLuisFernandoSepulveda.pdfapplication/pdf6098251http://tedebc.ufma.br:8080/bitstream/tede/2934/2/LuisFernandoSepulveda.pdf4e29a071a0a57272848ee7858986fdc0MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-82255http://tedebc.ufma.br:8080/bitstream/tede/2934/1/license.txt97eeade1fce43278e63fe063657f8083MD51tede/29342019-11-27 10:45:51.793oai:tede2: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Biblioteca Digital de Teses e Dissertaçõeshttps://tedebc.ufma.br/jspui/PUBhttp://tedebc.ufma.br:8080/oai/requestrepositorio@ufma.br||repositorio@ufma.bropendoar:21312019-11-27T13:45:51Biblioteca Digital de Teses e Dissertações da UFMA - Universidade Federal do Maranhão (UFMA)false |
dc.title.por.fl_str_mv |
Meta-learning applications in digital image processing |
dc.title.alternative.por.fl_str_mv |
Aplicações de meta-aprendizado no processamento de imagens digitais |
title |
Meta-learning applications in digital image processing |
spellingShingle |
Meta-learning applications in digital image processing SEPULVEDA, Luis Fernando Marin Meta-learning Image processing Machine-learning Feature selection Meta-data CNN Meta-aprendizagem Processamento de imagem Aprendizado de máquina Seleção de recursos Meta-dados CNN Engenharia Elétrica |
title_short |
Meta-learning applications in digital image processing |
title_full |
Meta-learning applications in digital image processing |
title_fullStr |
Meta-learning applications in digital image processing |
title_full_unstemmed |
Meta-learning applications in digital image processing |
title_sort |
Meta-learning applications in digital image processing |
author |
SEPULVEDA, Luis Fernando Marin |
author_facet |
SEPULVEDA, Luis Fernando Marin |
author_role |
author |
dc.contributor.advisor1.fl_str_mv |
SILVA, Aristófanes Corrêa |
dc.contributor.advisor1ID.fl_str_mv |
288745363-72 |
dc.contributor.advisor1Lattes.fl_str_mv |
http://lattes.cnpq.br/2446301582459104 |
dc.contributor.referee1.fl_str_mv |
SILVA, Aristófanes Corrêa |
dc.contributor.referee1ID.fl_str_mv |
288745363-72 |
dc.contributor.referee1Lattes.fl_str_mv |
http://lattes.cnpq.br/2446301582459104 |
dc.contributor.referee2.fl_str_mv |
CONCI, Aura |
dc.contributor.referee2ID.fl_str_mv |
https://orcid.org/0000-0003-0782-2501 |
dc.contributor.referee2Lattes.fl_str_mv |
http://lattes.cnpq.br/5601388085745497 |
dc.contributor.referee3.fl_str_mv |
LOPES, Denivaldo Cicero Pavao |
dc.contributor.referee3Lattes.fl_str_mv |
http://lattes.cnpq.br/7611180871627212 |
dc.contributor.referee4.fl_str_mv |
ALMEIDA, João Dallyson Sousa de |
dc.contributor.referee4Lattes.fl_str_mv |
http://lattes.cnpq.br/6047330108382641 |
dc.contributor.authorID.fl_str_mv |
629735713-75 |
dc.contributor.author.fl_str_mv |
SEPULVEDA, Luis Fernando Marin |
contributor_str_mv |
SILVA, Aristófanes Corrêa SILVA, Aristófanes Corrêa CONCI, Aura LOPES, Denivaldo Cicero Pavao ALMEIDA, João Dallyson Sousa de |
dc.subject.eng.fl_str_mv |
Meta-learning Image processing Machine-learning Feature selection Meta-data CNN |
topic |
Meta-learning Image processing Machine-learning Feature selection Meta-data CNN Meta-aprendizagem Processamento de imagem Aprendizado de máquina Seleção de recursos Meta-dados CNN Engenharia Elétrica |
dc.subject.por.fl_str_mv |
Meta-aprendizagem Processamento de imagem Aprendizado de máquina Seleção de recursos Meta-dados CNN |
dc.subject.cnpq.fl_str_mv |
Engenharia Elétrica |
description |
In recent decades, advances in capture devices and increase of available digital image data have stimulated the creation of methodologies for data processing that produce various forms of valuable models, such as descriptors, classifiers, approximations and visualizations. These models are often developed in the field of machine learning, which is characterized by a large number of available algorithms, these algorithms often do not have guidelines to identify the most appropriate one based on specific data to which they will be applied and nature of problem under analysis. There is a knowledge that allows to relate the features of the algorithms and data that present a good performance to fulfill a specific task, known as Meta-Knowledge, which can include information on algorithms, evaluation metrics to calculate similarity of datasets or relation of tasks. Being Meta-Learning the study of methods based on principles that explore the Meta-Knowledge to obtain efficient models and solutions, adapting the processes of Machine Learning and Data Mining. The research carried out in this work analyzes the applications and advantages offered by Meta-Learning in field of digital image processing. To carry out this task, different types of images, characterizers, and feature analysis techniques are used; in addition, multiple Machine Learning techniques are applied. The results obtained show that methodology based on Meta-Learning is efficient when applied in processing of digital images for identification and storage of experience generated by developing methodologies for classification of different types of images, obtaining a high performance with respect to an evaluation metrics. This statement means that Meta-Learning allows recommending the most appropriate methodology to perform the processing of a specific type of image based on features of dataset under analysis and the type of specific task to be performed. |
publishDate |
2019 |
dc.date.accessioned.fl_str_mv |
2019-11-27T13:45:51Z |
dc.date.issued.fl_str_mv |
2019-11-08 |
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.citation.fl_str_mv |
SEPULVEDA, Luis Fernando Marin. Meta-learning applications in digital image processing. 2019. 112 f. Dissertação (Programa de Pós-Graduação em Engenharia de Eletricidade/CCET) - Universidade Federal do Maranhão, São Luís, 2019 . |
dc.identifier.uri.fl_str_mv |
https://tedebc.ufma.br/jspui/handle/tede/tede/2934 |
identifier_str_mv |
SEPULVEDA, Luis Fernando Marin. Meta-learning applications in digital image processing. 2019. 112 f. Dissertação (Programa de Pós-Graduação em Engenharia de Eletricidade/CCET) - Universidade Federal do Maranhão, São Luís, 2019 . |
url |
https://tedebc.ufma.br/jspui/handle/tede/tede/2934 |
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.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal do Maranhão |
dc.publisher.program.fl_str_mv |
PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE ELETRICIDADE/CCET |
dc.publisher.initials.fl_str_mv |
UFMA |
dc.publisher.country.fl_str_mv |
Brasil |
dc.publisher.department.fl_str_mv |
DEPARTAMENTO DE ENGENHARIA DA ELETRICIDADE/CCET |
publisher.none.fl_str_mv |
Universidade Federal do Maranhão |
dc.source.none.fl_str_mv |
reponame:Biblioteca Digital de Teses e Dissertações da UFMA instname:Universidade Federal do Maranhão (UFMA) instacron:UFMA |
instname_str |
Universidade Federal do Maranhão (UFMA) |
instacron_str |
UFMA |
institution |
UFMA |
reponame_str |
Biblioteca Digital de Teses e Dissertações da UFMA |
collection |
Biblioteca Digital de Teses e Dissertações da UFMA |
bitstream.url.fl_str_mv |
http://tedebc.ufma.br:8080/bitstream/tede/2934/2/LuisFernandoSepulveda.pdf http://tedebc.ufma.br:8080/bitstream/tede/2934/1/license.txt |
bitstream.checksum.fl_str_mv |
4e29a071a0a57272848ee7858986fdc0 97eeade1fce43278e63fe063657f8083 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Biblioteca Digital de Teses e Dissertações da UFMA - Universidade Federal do Maranhão (UFMA) |
repository.mail.fl_str_mv |
repositorio@ufma.br||repositorio@ufma.br |
_version_ |
1800303798974939136 |