VisGraphNet: A complex network interpretation of convolutional neural features

Autor(es)

Florindo, Joao B. ^[1] ; Lee, Young-Sup ^[2] ; Jun, Kyungkoo ^[2] ; Jeon, Gwanggil ^[2] ; Albertini, Marcelo K. ^[3]

Afiliação do(s) autor(es)

^[1] Univ Estadual Campinas, Inst Math Stat & Sci Comp, Rua Sergio Buarque de Holanda 651, BR-13083859 Campinas, SP - Brazil

^[2] Incheon Natl Univ, Dept Embedded Syst Engn, 119 Acad Ro, Incheon 22012 - South Korea

^[3] Univ Fed Uberlandia, Dept Comp Sci, Av Joao Naves de Avila 2121, Room 1B150, Uberlandia, MG - Brazil

• Tipo de documento: Artigo Científico
• Fonte: INFORMATION SCIENCES; v. 543, p. 296-308, JAN 8 2021.
• Citações Web of Science: 0

Resumo

We propose and investigate the use of visibility graphs to model the feature map of a neural network. Initially devised for studies on complex networks, we employ this type of model for classification of texture images. An alternative viewpoint provided by these graphs over the original data motivates this work. Experiments evaluate the performance of our method using four benchmark databases, namely, KTHTIPS-2b, FMD, UIUC, and UMD and in a practical problem, which is the identification of plant species using scanned images of their leaves. Our method was competitive with other state-of-the-art approaches both in terms of classification accuracy and computational time. Results confirm the potential of techniques used for data analysis in different contexts to give more meaningful interpretation to the use of neural networks in texture classification. (C) 2020 Elsevier Inc. All rights reserved. (AU)

• Processo FAPESP: 16/16060-0 - Reconhecimento de Padrões em Imagens Baseado em Sistemas Complexos
• Beneficiário: Joao Batista Florindo
• Modalidade de apoio: Auxílio à Pesquisa - Regular