Semi-supervised learning with convolutional neural networks for UAV images automatic recognition

Detalhes bibliográficos
Autor(a) principal: Amorim, Willian Paraguassu
Data de Publicação: 2019
Outros Autores: Tetila, Everton Castelao, Pistori, Hemerson, Papa, Joao Paulo [UNESP]
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://dx.doi.org/10.1016/j.compag.2019.104932
http://hdl.handle.net/11449/186844
Resumo: The annotation of large datasets is an issue whose challenge increases as the number of labeled samples available to train the classifier reduces in comparison to the amount of unlabeled data. In this context, semi-supervised learning methods aim at discovering and propagating labels to unsupervised samples, such that their correct labeling can improve the classification performance. Our proposal makes use of semi-supervised methodologies to classify an unlabeled training set that is used to train a Convolution Neural Network using different training strategies. The proposed approach is experimentally validated for soybean leaf and herbivorous pest identification using images captured by Unmanned Aerial Vehicles and can support specialists and farmers in the pest control management in soybean fields, especially when they have a limited amount of labeled samples.
id UNSP_27ab7415ccc315fe00b196ccf62f029e
oai_identifier_str oai:repositorio.unesp.br:11449/186844
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling Semi-supervised learning with convolutional neural networks for UAV images automatic recognitionSemi-supervised learningConvolutional Neural NetworksFine tuningTransfer learningThe annotation of large datasets is an issue whose challenge increases as the number of labeled samples available to train the classifier reduces in comparison to the amount of unlabeled data. In this context, semi-supervised learning methods aim at discovering and propagating labels to unsupervised samples, such that their correct labeling can improve the classification performance. Our proposal makes use of semi-supervised methodologies to classify an unlabeled training set that is used to train a Convolution Neural Network using different training strategies. The proposed approach is experimentally validated for soybean leaf and herbivorous pest identification using images captured by Unmanned Aerial Vehicles and can support specialists and farmers in the pest control management in soybean fields, especially when they have a limited amount of labeled samples.Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Fundação para o Desenvolvimento da UNESP (FUNDUNESP)NVIDIA CorporationFoundation to Support the Development of Teaching, Science and Technology of the state of Mato Grosso do Sul (FUNDECT)Fed Univ Grande Dourados, BR-79804970 Dourados, BrazilUniv Catolica Dom Bosco, BR-79117900 Campo Grande, BrazilSao Paulo State Univ UNESP, BR-17033360 Bauru, BrazilSao Paulo State Univ UNESP, BR-17033360 Bauru, BrazilCNPq: 427968/2018-6CNPq: 307066/2017-7FAPESP: 2013/07375-0FAPESP: 2014/12236-1FAPESP: 2015/25739-4FAPESP: 2016/19403-6FUNDUNESP: 2597.2017Elsevier B.V.Fed Univ Grande DouradosUniv Catolica Dom BoscoUniversidade Estadual Paulista (Unesp)Amorim, Willian ParaguassuTetila, Everton CastelaoPistori, HemersonPapa, Joao Paulo [UNESP]2019-10-06T08:12:05Z2019-10-06T08:12:05Z2019-09-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article9http://dx.doi.org/10.1016/j.compag.2019.104932Computers And Electronics In Agriculture. Oxford: Elsevier Sci Ltd, v. 164, 9 p., 2019.0168-1699http://hdl.handle.net/11449/18684410.1016/j.compag.2019.104932WOS:000483910100030Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengComputers And Electronics In Agricultureinfo:eu-repo/semantics/openAccess2024-04-23T16:11:00Zoai:repositorio.unesp.br:11449/186844Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T21:00:32.926967Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
title Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
spellingShingle Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
Amorim, Willian Paraguassu
Semi-supervised learning
Convolutional Neural Networks
Fine tuning
Transfer learning
title_short Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
title_full Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
title_fullStr Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
title_full_unstemmed Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
title_sort Semi-supervised learning with convolutional neural networks for UAV images automatic recognition
author Amorim, Willian Paraguassu
author_facet Amorim, Willian Paraguassu
Tetila, Everton Castelao
Pistori, Hemerson
Papa, Joao Paulo [UNESP]
author_role author
author2 Tetila, Everton Castelao
Pistori, Hemerson
Papa, Joao Paulo [UNESP]
author2_role author
author
author
dc.contributor.none.fl_str_mv Fed Univ Grande Dourados
Univ Catolica Dom Bosco
Universidade Estadual Paulista (Unesp)
dc.contributor.author.fl_str_mv Amorim, Willian Paraguassu
Tetila, Everton Castelao
Pistori, Hemerson
Papa, Joao Paulo [UNESP]
dc.subject.por.fl_str_mv Semi-supervised learning
Convolutional Neural Networks
Fine tuning
Transfer learning
topic Semi-supervised learning
Convolutional Neural Networks
Fine tuning
Transfer learning
description The annotation of large datasets is an issue whose challenge increases as the number of labeled samples available to train the classifier reduces in comparison to the amount of unlabeled data. In this context, semi-supervised learning methods aim at discovering and propagating labels to unsupervised samples, such that their correct labeling can improve the classification performance. Our proposal makes use of semi-supervised methodologies to classify an unlabeled training set that is used to train a Convolution Neural Network using different training strategies. The proposed approach is experimentally validated for soybean leaf and herbivorous pest identification using images captured by Unmanned Aerial Vehicles and can support specialists and farmers in the pest control management in soybean fields, especially when they have a limited amount of labeled samples.
publishDate 2019
dc.date.none.fl_str_mv 2019-10-06T08:12:05Z
2019-10-06T08:12:05Z
2019-09-01
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://dx.doi.org/10.1016/j.compag.2019.104932
Computers And Electronics In Agriculture. Oxford: Elsevier Sci Ltd, v. 164, 9 p., 2019.
0168-1699
http://hdl.handle.net/11449/186844
10.1016/j.compag.2019.104932
WOS:000483910100030
url http://dx.doi.org/10.1016/j.compag.2019.104932
http://hdl.handle.net/11449/186844
identifier_str_mv Computers And Electronics In Agriculture. Oxford: Elsevier Sci Ltd, v. 164, 9 p., 2019.
0168-1699
10.1016/j.compag.2019.104932
WOS:000483910100030
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Computers And Electronics In Agriculture
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 9
dc.publisher.none.fl_str_mv Elsevier B.V.
publisher.none.fl_str_mv Elsevier B.V.
dc.source.none.fl_str_mv Web of Science
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1808129273481920512