Effective Defect Detection Method Based on Bilinear Texture Features for LGPs

Automatic defect detection of light guide plates (LGPs) is an important task in the manufacture of liquid crystal displays. During thermo-printing, defects of tag lines on LGPs may occur easily, and these defects are of two categories: bubbles and missing tag lines. These defects lack salient visual...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Libin Hong, Xianglei Wu, Dibin Zhou, Fuchang Liu
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Defects detection texture classification bilinear convolutional neural networks Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/2836c72c54954418a6ed8e0421d98d93
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2836c72c54954418a6ed8e0421d98d93
record_format	dspace
spelling	oai:doaj.org-article:2836c72c54954418a6ed8e0421d98d932021-11-18T00:05:05ZEffective Defect Detection Method Based on Bilinear Texture Features for LGPs2169-353610.1109/ACCESS.2021.3111410https://doaj.org/article/2836c72c54954418a6ed8e0421d98d932021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9531362/https://doaj.org/toc/2169-3536Automatic defect detection of light guide plates (LGPs) is an important task in the manufacture of liquid crystal displays. During thermo-printing, defects of tag lines on LGPs may occur easily, and these defects are of two categories: bubbles and missing tag lines. These defects lack salient visual attributes, such as edge-based and region-based features, and as such, traditional methods fail to detect them. To address this, we propose a Dense-bilinear convolutional neural network (BCNN), an end-to-end defect detection network, utilizing Dense-blocks (Huang <italic>et al.</italic>, 2017), Bilinear feature layers (Lin <italic>et al.</italic>, 2015), and squeeze-and-excitation blocks (Hu <italic>et al.</italic>, 2018). Our network exploits fine-grained texture features, which leads to parameter reduction and accuracy enhancement. We validate our network on our LGP dataset containing 5,860 images from three cases: bubbles, tag line existence, and tag line missing. Our network outperforms AlexNet (Krizhevsky <italic>et al.</italic>, 2012), VGG (Simonyan and Zisserman, 2014) and ResNet (He <italic>et al.</italic>, 2016), on both the public and our LGP datasets with less GPU memory consumption.Libin HongXianglei WuDibin ZhouFuchang LiuIEEEarticleDefects detectiontexture classificationbilinear convolutional neural networksElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 147958-147966 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Defects detection texture classification bilinear convolutional neural networks Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Defects detection texture classification bilinear convolutional neural networks Electrical engineering. Electronics. Nuclear engineering TK1-9971 Libin Hong Xianglei Wu Dibin Zhou Fuchang Liu Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
description	Automatic defect detection of light guide plates (LGPs) is an important task in the manufacture of liquid crystal displays. During thermo-printing, defects of tag lines on LGPs may occur easily, and these defects are of two categories: bubbles and missing tag lines. These defects lack salient visual attributes, such as edge-based and region-based features, and as such, traditional methods fail to detect them. To address this, we propose a Dense-bilinear convolutional neural network (BCNN), an end-to-end defect detection network, utilizing Dense-blocks (Huang <italic>et al.</italic>, 2017), Bilinear feature layers (Lin <italic>et al.</italic>, 2015), and squeeze-and-excitation blocks (Hu <italic>et al.</italic>, 2018). Our network exploits fine-grained texture features, which leads to parameter reduction and accuracy enhancement. We validate our network on our LGP dataset containing 5,860 images from three cases: bubbles, tag line existence, and tag line missing. Our network outperforms AlexNet (Krizhevsky <italic>et al.</italic>, 2012), VGG (Simonyan and Zisserman, 2014) and ResNet (He <italic>et al.</italic>, 2016), on both the public and our LGP datasets with less GPU memory consumption.
format	article
author	Libin Hong Xianglei Wu Dibin Zhou Fuchang Liu
author_facet	Libin Hong Xianglei Wu Dibin Zhou Fuchang Liu
author_sort	Libin Hong
title	Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
title_short	Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
title_full	Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
title_fullStr	Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
title_full_unstemmed	Effective Defect Detection Method Based on Bilinear Texture Features for LGPs
title_sort	effective defect detection method based on bilinear texture features for lgps
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/2836c72c54954418a6ed8e0421d98d93
work_keys_str_mv	AT libinhong effectivedefectdetectionmethodbasedonbilineartexturefeaturesforlgps AT xiangleiwu effectivedefectdetectionmethodbasedonbilineartexturefeaturesforlgps AT dibinzhou effectivedefectdetectionmethodbasedonbilineartexturefeaturesforlgps AT fuchangliu effectivedefectdetectionmethodbasedonbilineartexturefeaturesforlgps
_version_	1718425213222256640

Effective Defect Detection Method Based on Bilinear Texture Features for LGPs

Ejemplares similares