A multi-layer feature fusion method for few-shot image classification

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversity of Brasília, Department of Mechanical Engineering-
Autor(es): dc.contributorEmbrapa Cerrados-
Autor(es): dc.contributorUniversity of Brasília, Department of Computer Science-
Autor(es): dc.creatorGomes, Jacó Cirino-
Autor(es): dc.creatorBorges, Lurdineide de Araújo Barbosa-
Autor(es): dc.creatorBorges, Díbio Leandro-
Data de aceite: dc.date.accessioned2026-08-11T11:00:54Z-
Data de disponibilização: dc.date.available2026-08-11T11:00:54Z-
Data de envio: dc.date.issued2025-09-19-
Data de envio: dc.date.issued2025-09-19-
Data de envio: dc.date.issued2023-08-03-
Fonte completa do material: dc.identifierhttp://repositorio.unb.br/handle/10482/52466-
Fonte completa do material: dc.identifierhttps://doi.org/10.3390/s23156880-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0003-4810-5138-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0001-9284-3299-
Fonte completa do material: dc.identifierhttps://orcid.org/0000-0002-4868-0629-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/capes/1186645-
Descrição: dc.descriptionIn image classification, few-shot learning deals with recognizing visual categories from a few tagged examples. The degree of expressiveness of the encoded features in this scenario is a crucial question that needs to be addressed in the models being trained. Recent approaches have achieved encouraging results in improving few-shot models in deep learning, but designing a competitive and simple architecture is challenging, especially considering its requirement in many practical applications. This work proposes an improved few-shot model based on a multi-layer feature fusion (FMLF) method. The presented approach includes extended feature extraction and fusion mechanisms in the Convolutional Neural Network (CNN) backbone, as well as an effective metric to compute the divergences in the end. In order to evaluate the proposed method, a challenging visual classification problem, maize crop insect classification with specific pests and beneficial categories, is addressed, serving both as a test of our model and as a means to propose a novel dataset. Experiments were carried out to compare the results with ResNet50, VGG16, and MobileNetv2, used as feature extraction backbones, and the FMLF method demonstrated higher accuracy with fewer parameters. The proposed FMLF method improved accuracy scores by up to 3.62% in one-shot and 2.82% in fiveshot classification tasks compared to a traditional backbone, which uses only global image features.-
Descrição: dc.descriptionInstituto de Ciências Exatas (IE)-
Descrição: dc.descriptionDepartamento de Ciência da Computação (IE CIC)-
Descrição: dc.descriptionPrograma de Pós-Graduação em Sistemas Mecatrônicos-
Formato: dc.formatapplication/pdf-
Idioma: dc.languageen-
Publicador: dc.publisherMDPI-
Direitos: dc.rightsAcesso Aberto-
Direitos: dc.rights(CC BY) © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).-
Palavras-chave: dc.subjectRede Neurais Convolucionais (CNNs)-
Palavras-chave: dc.subjectClassificação de imagens-
Palavras-chave: dc.subjectMultiescala-
Palavras-chave: dc.subjectAprendizagem métrica-
Título: dc.titleA multi-layer feature fusion method for few-shot image classification-
Tipo de arquivo: dc.typelivro digital-
Aparece nas coleções:Repositório Institucional – UNB

Não existem arquivos associados a este item.