Deep convolutional networks based on lightweight YOLOv8 to detect and estimate peanut losses from images in post-harvesting environments

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.contributorFederal Technological University of Paraná (UTFPR)-
Autor(es): dc.contributorKansas State University-
Autor(es): dc.contributorUniversity of Georgia-
Autor(es): dc.creatorBrito Filho, Armando Lopes de-
Autor(es): dc.creatorMorlin Carneiro, Franciele-
Autor(es): dc.creatorCarreira, Vinicius dos Santos-
Autor(es): dc.creatorTedesco, Danilo-
Autor(es): dc.creatorCosta Souza, Jarlyson Brunno-
Autor(es): dc.creatorBarbosa Júnior, Marcelo Rodrigues-
Autor(es): dc.creatorSilva, Rouverson Pereira da-
Data de aceite: dc.date.accessioned2025-08-21T17:29:03Z-
Data de disponibilização: dc.date.available2025-08-21T17:29:03Z-
Data de envio: dc.date.issued2025-04-29-
Data de envio: dc.date.issued2025-07-01-
Fonte completa do material: dc.identifierhttp://dx.doi.org/10.1016/j.compag.2025.110282-
Fonte completa do material: dc.identifierhttps://hdl.handle.net/11449/299546-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/299546-
Descrição: dc.descriptionPeanut losses detection is key to monitor operational quality during mechanical harvesting. Current manual assessments faces practical limitations in the field, as they tend to be exhaustive, time-consuming, and susceptible to errors, especially after long work periods. Therefore, the main objective of this study was to develop an automated image processing framework to detect, count, and estimate peanut pod losses during the harvesting operation. We proposed a robust approach encompassing different environmental conditions and training detection algorithms, specifically based on lightweight YOLOv8 architecture, with images acquired with a mobile smartphone at six different times of the day (10 a.m., 11 a.m., 1 p.m., 2 p.m., 3 p.m., and 4 p.m.). The experimental results showed that detecting two-seed peanut pods was more effective than one-seed pods, with higher precision, recall, and mAP50 values. The best results for image acquisition were between 10 a.m. and 2 p.m. The study also compared manual and automated counting methods, revealing that the best scenarios for counting achieved an R2 above 0.80. Furthermore, georeferenced maps of peanut losses revealed significant spatial variability, providing critical insights for targeted interventions. These findings demonstrate the potential to enhance mechanized harvesting efficiency and lay the groundwork for future integration into fully automated systems. By incorporating this method into harvesting machinery, real-time monitoring and accurate loss quantification can be achieved, substantially reducing the need for labor-intensive manual assessments.-
Descrição: dc.descriptionDepartment of Engineering and Mathematical Sciences School of Agricultural and Veterinarian Sciences São Paulo State University (Unesp) Jaboticabal-
Descrição: dc.descriptionFederal Technological University of Paraná (UTFPR), Paraná-
Descrição: dc.descriptionDepartment of Agronomy Kansas State University-
Descrição: dc.descriptionDepartment of Horticulture University of Georgia-
Descrição: dc.descriptionDepartment of Engineering and Mathematical Sciences School of Agricultural and Veterinarian Sciences São Paulo State University (Unesp) Jaboticabal-
Idioma: dc.languageen-
Relação: dc.relationComputers and Electronics in Agriculture-
???dc.source???: dc.sourceScopus-
Palavras-chave: dc.subjectArachis hypogaea L-
Palavras-chave: dc.subjectObject detection-
Palavras-chave: dc.subjectPeanut losses-
Palavras-chave: dc.subjectSmart harvesting-
Palavras-chave: dc.subjectYOLOv8-
Título: dc.titleDeep convolutional networks based on lightweight YOLOv8 to detect and estimate peanut losses from images in post-harvesting environments-
Tipo de arquivo: dc.typelivro digital-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.