Logo do repositório
 
Publicação

Impact of hyper-parameter tuning on CNN accuracy in agricultural image classification

2025Artigo científico Acesso aberto
Ver registo simples
datacite.subject.fosCiências Agrárias::Agricultura, Silvicultura e Pescas
datacite.subject.fosEngenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
datacite.subject.sdg04:Educação de Qualidade
datacite.subject.sdg09:Indústria, Inovação e Infraestruturas
datacite.subject.sdg12:Produção e Consumo Sustentáveis
dc.contributor.authorMendes, João
dc.contributor.authorLima, José
dc.contributor.authorCosta, Lino
dc.contributor.authorHendrix, Eligius M.T.
dc.contributor.authorPereira, Ana I.
dc.date.accessioned2025-07-10T09:27:27Z
dc.date.available2025-07-10T09:27:27Z
dc.date.issued2025
dc.description.abstractThis study explores the impact of hyper-parameter optimization on the performance of convolutional neural networks (CNNs) for olive cultivar classification using transfer learning. Pre-trained ImageNet models such as VGG16, InceptionV3, and ResNet50 were adapted to a proprietary dataset, with VGG16 selected for detailed evaluation. Key hyper-parameters, including layer count, neurons per layer, dropout rate, learning rate, and batch size, were tuned using random search. The best configuration achieved a validation accuracy of 87.5%, significantly outperforming the control model. Sensitivity analyses with Morris and Sobol methods identified the number of layers as the most influential factor, followed by dropout and learning rates through interaction effects. These findings demonstrate the importance of tailoring CNN architecture and regularization settings to the problem domain. These results underscore the importance of tuning architectural depth and regularization mechanisms for performance optimization. As a practical guideline, models with fewer layers and intermediate dropout levels demonstrated higher robustness and generalization, offering an effective strategy for adapting CNNs to agricultural classification tasks.eng
dc.description.sponsorshipThis work was supported by national funds through FCT/MCTES (PIDDAC): CeDRI, UID/05757 (DOI: https://doi.org/10.54499/UIDB/05757/2020 and DOI: https://doi.org/10.54499/UIDP/05757/2020), SusTEC, LA/P/0007/2020 (DOI: https://doi.org/10.54499/LA/P/0007/2020) and Algoritmi UIDB/00319/2020.
dc.identifier.citationMendes, João; Lima, José; Costa, Lino; Hendrix, Eligius M.T.; Pereira, Ana I. (2025). Impact of hyper-parameter tuning on CNN accuracy in agricultural image classification. Smart Agricultural Technology. ISSN 2772-3755.
dc.identifier.doi10.1016/j.atech.2025.101016
dc.identifier.issn2772-3755
dc.identifier.urihttp://hdl.handle.net/10198/34657
dc.language.isoeng
dc.peerreviewedyes
dc.publisherElsevier BV
dc.relationResearch Centre in Digitalization and Intelligent Robotics
dc.relationResearch Centre in Digitalization and Intelligent Robotics
dc.relationAssociate Laboratory for Sustainability and Tecnology in Mountain Regions
dc.relationALGORITMI Research Center
dc.relation.ispartofSmart Agricultural Technology
dc.rights.urihttp://creativecommons.org/licenses/by-nd/4.0/
dc.subjectHyper-parameter optimization
dc.subjectConvolutional neural networks
dc.subjectSensitivity analysis
dc.titleImpact of hyper-parameter tuning on CNN accuracy in agricultural image classification
dc.typejournal article
dcterms.referenceshttps://doi.org/10.34620/dadosipb/TVYE8K
dspace.entity.typePublication
oaire.awardTitleResearch Centre in Digitalization and Intelligent Robotics
oaire.awardTitleResearch Centre in Digitalization and Intelligent Robotics
oaire.awardTitleAssociate Laboratory for Sustainability and Tecnology in Mountain Regions
oaire.awardTitleALGORITMI Research Center
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F05757%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F05757%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/LA%2FP%2F0007%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00319%2F2020/PT
oaire.citation.startPage1
oaire.citation.titleSmart Agricultural Technology
oaire.citation.volume11
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameMendes
person.familyNameLima
person.familyNamePereira
person.givenNameJoão
person.givenNameJosé
person.givenNameAna I.
person.identifier2726655
person.identifierR-000-8GD
person.identifier.ciencia-idEA1F-844D-6BA9
person.identifier.ciencia-id6016-C902-86A9
person.identifier.ciencia-id0716-B7C2-93E4
person.identifier.orcid0000-0003-0979-8314
person.identifier.orcid0000-0001-7902-1207
person.identifier.orcid0000-0003-3803-2043
person.identifier.ridL-3370-2014
person.identifier.ridF-3168-2010
person.identifier.scopus-author-id57225794972
person.identifier.scopus-author-id55851941311
person.identifier.scopus-author-id15071961600
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
relation.isAuthorOfPublicationb5c9de22-cf9e-47b8-b7a4-26e08fb12b28
relation.isAuthorOfPublicationd88c2b2a-efc2-48ef-b1fd-1145475e0055
relation.isAuthorOfPublicatione9981d62-2a2b-4fef-b75e-c2a14b0e7846
relation.isAuthorOfPublication.latestForDiscoveryb5c9de22-cf9e-47b8-b7a4-26e08fb12b28
relation.isProjectOfPublication6e01ddc8-6a82-4131-bca6-84789fa234bd
relation.isProjectOfPublicationd0a17270-80a8-4985-9644-a04c2a9f2dff
relation.isProjectOfPublication6255046e-bc79-4b82-8884-8b52074b4384
relation.isProjectOfPublication0d98f999-8fd3-46a8-8a71-a7ff478a1207
relation.isProjectOfPublication.latestForDiscovery6e01ddc8-6a82-4131-bca6-84789fa234bd

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
1-s2.0-S2772375525002497-main.pdf
Tamanho:
3.12 MB
Formato:
Adobe Portable Document Format
Ver/Abrir
Licença
A mostrar 1 - 1 de 1
Miniatura indisponível
Nome:
license.txt
Tamanho:
1.75 KB
Formato:
Item-specific license agreed upon to submission
Descrição:
Ver/Abrir

Coleções

ESTiG - Artigos em Revistas Indexados à WoS/Scopus