Cargando…

Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM

A convolutional neural network (CNN), which exhibits excellent performance in solving image-based problem, has been widely applied to various industrial problems. In general, the CNN model was applied to defect inspection on the surface of raw materials or final products, and its accuracy also showe...

Descripción completa

Detalles Bibliográficos
Autores principales: Moon, In Yong, Lee, Ho Won, Kim, Se-Jong, Oh, Young-Seok, Jung, Jaimyun, Kang, Seong-Hoon
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8122604/
https://www.ncbi.nlm.nih.gov/pubmed/33919231
http://dx.doi.org/10.3390/ma14092095
_version_ 1783692662556065792
author Moon, In Yong
Lee, Ho Won
Kim, Se-Jong
Oh, Young-Seok
Jung, Jaimyun
Kang, Seong-Hoon
author_facet Moon, In Yong
Lee, Ho Won
Kim, Se-Jong
Oh, Young-Seok
Jung, Jaimyun
Kang, Seong-Hoon
author_sort Moon, In Yong
collection PubMed
description A convolutional neural network (CNN), which exhibits excellent performance in solving image-based problem, has been widely applied to various industrial problems. In general, the CNN model was applied to defect inspection on the surface of raw materials or final products, and its accuracy also showed better performance compared to human inspection. However, surfaces with heterogeneous and complex backgrounds have difficulties in separating defects region from the background, which is a typical challenge in this field. In this study, the CNN model was applied to detect surface defects on a hierarchical patterned surface, one of the representative complex background surfaces. In order to optimize the CNN structure, the change in inspection performance was analyzed according to the number of layers and kernel size of the model using evaluation metrics. In addition, the change of the CNN’s decision criteria according to the change of the model structure was analyzed using a class activation map (CAM) technique, which can highlight the most important region recognized by the CNN in performing classification. As a result, we were able to accurately understand the classification manner of the CNN for the hierarchical pattern surface, and an accuracy of 93.7% was achieved using the optimized model.
format Online
Article
Text
id pubmed-8122604
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-81226042021-05-16 Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM Moon, In Yong Lee, Ho Won Kim, Se-Jong Oh, Young-Seok Jung, Jaimyun Kang, Seong-Hoon Materials (Basel) Article A convolutional neural network (CNN), which exhibits excellent performance in solving image-based problem, has been widely applied to various industrial problems. In general, the CNN model was applied to defect inspection on the surface of raw materials or final products, and its accuracy also showed better performance compared to human inspection. However, surfaces with heterogeneous and complex backgrounds have difficulties in separating defects region from the background, which is a typical challenge in this field. In this study, the CNN model was applied to detect surface defects on a hierarchical patterned surface, one of the representative complex background surfaces. In order to optimize the CNN structure, the change in inspection performance was analyzed according to the number of layers and kernel size of the model using evaluation metrics. In addition, the change of the CNN’s decision criteria according to the change of the model structure was analyzed using a class activation map (CAM) technique, which can highlight the most important region recognized by the CNN in performing classification. As a result, we were able to accurately understand the classification manner of the CNN for the hierarchical pattern surface, and an accuracy of 93.7% was achieved using the optimized model. MDPI 2021-04-21 /pmc/articles/PMC8122604/ /pubmed/33919231 http://dx.doi.org/10.3390/ma14092095 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/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/).
spellingShingle Article
Moon, In Yong
Lee, Ho Won
Kim, Se-Jong
Oh, Young-Seok
Jung, Jaimyun
Kang, Seong-Hoon
Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title_full Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title_fullStr Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title_full_unstemmed Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title_short Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM
title_sort analysis of the region of interest according to cnn structure in hierarchical pattern surface inspection using cam
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8122604/
https://www.ncbi.nlm.nih.gov/pubmed/33919231
http://dx.doi.org/10.3390/ma14092095
work_keys_str_mv AT mooninyong analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam
AT leehowon analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam
AT kimsejong analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam
AT ohyoungseok analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam
AT jungjaimyun analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam
AT kangseonghoon analysisoftheregionofinterestaccordingtocnnstructureinhierarchicalpatternsurfaceinspectionusingcam