Art

J-GLOBAL ID：202102246643235536   Reference number：21A0449973

Detection of minute defects using transfer learning-based CNN models

転移学習ベースCNNモデルを用いた微小欠陥の検出

Author (7)： Nakashima Kento (Graduate School of Engineering, Sanyo-Onoda City University, Sanyo-Onoda, Japan) ,  Nagata Fusaomi (Graduate School of Engineering, Sanyo-Onoda City University, Sanyo-Onoda, Japan) ,  Ochi Hiroaki (Graduate School of Engineering, Sanyo-Onoda City University, Sanyo-Onoda, Japan) ,  Otsuka Akimasa (Graduate School of Engineering, Sanyo-Onoda City University, Sanyo-Onoda, Japan) ,  Ikeda Takeshi (Graduate School of Engineering, Sanyo-Onoda City University, Sanyo-Onoda, Japan) ,  Watanabe Keigo (Okayama University, Okayama, Japan) ,  Habib Maki K. (American University in Cairo, Cairo, Egypt)
Material： Artificial Life and Robotics
Volume： 26  Issue： 1  Page： 35-41  Publication year： Feb. 2021 
JST Material Number： L3251A  ISSN： 1433-5298  Document type： Article
Article type： 原著論文  Country of issue： Germany, Federal Republic of (DEU)  Language： ENGLISH (EN)

Thesaurus term： *Machine Learning ,  *neural network (biomodel) ,  plastics molding ,  *fault detection ,  directed graph ,  state ,  information ,  classification ,  convolutional neuralnetwork ,  binary classification
Semi thesaurus term： *convolutional neural network ,  Binary classification ,  image information ,  *fault detection ,  resin molding ,  series ,  *transfer learning ,  directed acyclic graph

Author keywords (4)： Convolutional neural network (CNN) ,  CNN design tool ,  Transfer learning ,  Minute defect detection

JST classification (2)： Pattern recognition  (JE07000S) ,  Artificial intelligence  (JE08000Z)

Terms in the title (5)： 転移学習 ,  CNN ,  モデル ,  微小 ,  欠陥