Mapping stains on flat roofs using semantic segmentation based on deep learning

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversity of Brasilia-
Autor(es): dc.contributorDom Bosco Catholic University-
Autor(es): dc.contributorDom Bosco Catholic University-
Autor(es): dc.contributorUniversity of Brasilia-
Autor(es): dc.contributorUniversity of Brasilia-
Autor(es): dc.contributorUniversity of Augsburg, Centre for Climate Resilience-
Autor(es): dc.contributorSalesian Polytechnic University, Environmental Research Group for Sustainable Development (GIADES)-
Autor(es): dc.contributorDom Bosco Catholic University-
Autor(es): dc.contributorFederal University of Mato Grosso do Sul-
Autor(es): dc.creatorSantos, Lara Monalisa Alves dos-
Autor(es): dc.creatorLescano, Leonardo Rabero-
Autor(es): dc.creatorHiga, Gabriel Toshio Hirokawa-
Autor(es): dc.creatorZanoni, Vanda Alice Garcia-
Autor(es): dc.creatorSilva, Lenildo Santos da-
Autor(es): dc.creatorAlvarez Mendoza, Cesar Ivan-
Autor(es): dc.creatorPistori, Hemerson-
Data de aceite: dc.date.accessioned2025-03-18T17:17:03Z-
Data de disponibilização: dc.date.available2025-03-18T17:17:03Z-
Data de envio: dc.date.issued2025-03-17-
Data de envio: dc.date.issued2025-03-17-
Data de envio: dc.date.issued2024-12-17-
Fonte completa do material: dc.identifierhttp://repositorio.unb.br/handle/10482/51900-
Fonte completa do material: dc.identifierhttps://doi.org/10.1016/j.cscm.2024.e04106-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0002-8022-2513-
Fonte completa do material: dc.identifierhttps://orcid.org/0009-0004-3125-9696-
Fonte completa do material: dc.identifierhttps://orcid.org/0009-0006-6771-0076-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0003-2629-4214-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0001-5099-6123-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0001-5629-0893-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0001-8181-760X-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/capes/925404-
Descrição: dc.descriptionMoisture stains indicate ongoing degradation processes and may reveal areas of the roof slab where water infiltration occurs, compromising the performance and durability of the building system. During inspections of roofing systems, an inspector’s field of vision differs from that of drones during overflights. As a result, traditional inspections might not always detect the presence and severity of stains, making maintenance on flat roofs a complex task. In this context, this experimental study aims to analyze deep learning-based semantic segmentation with images obtained from drones to map and monitor damp patches during automated building inspections of flat roof systems. The research tested two convolutional neural networks for semantic segmentation: the Fully Convolutional Network (FCN) with a ResNet50 backbone and DeepLabV3 with a ResNet101 backbone, as well as a transformer-based deep artificial neural network called SegFormer with a MiT-B1 backbone. We evaluated three optimizers for each model—Adam, Adagrad, and SGD—along with learning rates of 1e-2, 1e-3, and 1e-4. The models were compared using four performance metrics. The FCN, optimized with Adagrad at a learning rate of 1e-2, showed the best results. The average metrics obtained in this case were as follows: precision: 79.69 %, recall: 67.81 %, F-score: 73.09 %, and Intersection over Union (IoU): 57.70 %.-
Descrição: dc.descriptionFaculdade de Arquitetura e Urbanismo (FAU)-
Descrição: dc.descriptionDepartamento de Tecnologia em Arquitetura e Urbanismo (FAU TEC)-
Descrição: dc.descriptionFaculdade de Tecnologia (FT)-
Descrição: dc.descriptionDepartamento de Engenharia Civil e Ambiental (FT ENC)-
Descrição: dc.descriptionPrograma de Pós-Graduação em Arquitetura e Urbanismo-
Formato: dc.formatapplication/pdf-
Idioma: dc.languageen-
Publicador: dc.publisherElsevier Ltd.-
Direitos: dc.rightsAcesso Aberto-
Direitos: dc.rightsThis is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).-
Palavras-chave: dc.subjectAprendizagem profunda-
Palavras-chave: dc.subjectInspeção predial-
Palavras-chave: dc.subjectDrones-
Palavras-chave: dc.subjectVisão computacional-
Título: dc.titleMapping stains on flat roofs using semantic segmentation based on deep learning-
Tipo de arquivo: dc.typelivro digital-
Aparece nas coleções:Repositório Institucional – UNB - Rep. 1

Não existem arquivos associados a este item.