Cascading Convolutional Neural Network for Steel Surface Defect Detection

Chih Yang Lin; Cheng Hsun Chen; Ching Yuan Yang; Fityanul Akhyar; Chao Yung Hsu; Hui Fuang Ng

doi:10.1007/978-3-030-20454-9_20

Cascading Convolutional Neural Network for Steel Surface Defect Detection

Chih Yang Lin, Cheng Hsun Chen, Ching Yuan Yang, Fityanul Akhyar, Chao Yung Hsu, Hui Fuang Ng

Department of Mechanical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

23 Scopus citations

Abstract

Steel is the most important material in the world of engineering and construction. Modern steelmaking relies on computer vision technologies, like optical cameras to monitor the production and manufacturing processes, which helps companies improve product quality. In this paper, we propose a deep learning method to automatically detect defects on the steel surface. The architecture of our proposed system is separated into two parts. The first part uses a revised version of single shot multibox detector (SSD) model to learn possible defects. Then, deep residual network (ResNet) is used to classify three types of defects: Rust, Scar, and Sponge. The combination of these two models is investigated and discussed thoroughly in this paper. This work additionally employs a real industry dataset to confirm the feasibility of the proposed method and make sure it is applicable to real-world scenarios. The experimental results show that the proposed method can achieve higher precision and recall scores in steel surface defect detection.

Original language	English
Title of host publication	Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019
Editors	Tareq Ahram
Publisher	Springer Verlag
Pages	202-212
Number of pages	11
ISBN (Print)	9783030204532
DOIs	https://doi.org/10.1007/978-3-030-20454-9_20
State	Published - 2020
Event	AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019 - Washington D.C., United States Duration: 24 Jul 2019 → 28 Jul 2019

Publication series

Name	Advances in Intelligent Systems and Computing
Volume	965
ISSN (Print)	2194-5357
ISSN (Electronic)	2194-5365

Conference

Conference	AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019
Country/Territory	United States
City	Washington D.C.
Period	24/07/19 → 28/07/19

Keywords

Defect detection
Fully convolutional networks
ResNet
SSD

Access to Document

10.1007/978-3-030-20454-9_20

The Intelligent Parter of the Blind( II )
Wang, W.-J. (PI)
1/01/19 → 31/12/19
Project: Research

Cite this

Lin, C. Y., Chen, C. H., Yang, C. Y., Akhyar, F., Hsu, C. Y., & Ng, H. F. (2020). Cascading Convolutional Neural Network for Steel Surface Defect Detection. In T. Ahram (Ed.), Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019 (pp. 202-212). (Advances in Intelligent Systems and Computing; Vol. 965). Springer Verlag. https://doi.org/10.1007/978-3-030-20454-9_20

Lin, Chih Yang ; Chen, Cheng Hsun ; Yang, Ching Yuan et al. / Cascading Convolutional Neural Network for Steel Surface Defect Detection. Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019. editor / Tareq Ahram. Springer Verlag, 2020. pp. 202-212 (Advances in Intelligent Systems and Computing).

@inproceedings{7e87f558e5cf4a0ca38a47820624cc8e,

title = "Cascading Convolutional Neural Network for Steel Surface Defect Detection",

abstract = "Steel is the most important material in the world of engineering and construction. Modern steelmaking relies on computer vision technologies, like optical cameras to monitor the production and manufacturing processes, which helps companies improve product quality. In this paper, we propose a deep learning method to automatically detect defects on the steel surface. The architecture of our proposed system is separated into two parts. The first part uses a revised version of single shot multibox detector (SSD) model to learn possible defects. Then, deep residual network (ResNet) is used to classify three types of defects: Rust, Scar, and Sponge. The combination of these two models is investigated and discussed thoroughly in this paper. This work additionally employs a real industry dataset to confirm the feasibility of the proposed method and make sure it is applicable to real-world scenarios. The experimental results show that the proposed method can achieve higher precision and recall scores in steel surface defect detection.",

keywords = "Defect detection, Fully convolutional networks, ResNet, SSD",

author = "Lin, {Chih Yang} and Chen, {Cheng Hsun} and Yang, {Ching Yuan} and Fityanul Akhyar and Hsu, {Chao Yung} and Ng, {Hui Fuang}",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019 ; Conference date: 24-07-2019 Through 28-07-2019",

year = "2020",

doi = "10.1007/978-3-030-20454-9_20",

language = "???core.languages.en_GB???",

isbn = "9783030204532",

series = "Advances in Intelligent Systems and Computing",

publisher = "Springer Verlag",

pages = "202--212",

editor = "Tareq Ahram",

booktitle = "Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019",

}

Lin, CY, Chen, CH, Yang, CY, Akhyar, F, Hsu, CY & Ng, HF 2020, Cascading Convolutional Neural Network for Steel Surface Defect Detection. in T Ahram (ed.), Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019. Advances in Intelligent Systems and Computing, vol. 965, Springer Verlag, pp. 202-212, AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019, Washington D.C., United States, 24/07/19. https://doi.org/10.1007/978-3-030-20454-9_20

Cascading Convolutional Neural Network for Steel Surface Defect Detection. / Lin, Chih Yang; Chen, Cheng Hsun; Yang, Ching Yuan et al.
Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019. ed. / Tareq Ahram. Springer Verlag, 2020. p. 202-212 (Advances in Intelligent Systems and Computing; Vol. 965).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Cascading Convolutional Neural Network for Steel Surface Defect Detection

AU - Lin, Chih Yang

AU - Chen, Cheng Hsun

AU - Yang, Ching Yuan

AU - Akhyar, Fityanul

AU - Hsu, Chao Yung

AU - Ng, Hui Fuang

PY - 2020

Y1 - 2020

N2 - Steel is the most important material in the world of engineering and construction. Modern steelmaking relies on computer vision technologies, like optical cameras to monitor the production and manufacturing processes, which helps companies improve product quality. In this paper, we propose a deep learning method to automatically detect defects on the steel surface. The architecture of our proposed system is separated into two parts. The first part uses a revised version of single shot multibox detector (SSD) model to learn possible defects. Then, deep residual network (ResNet) is used to classify three types of defects: Rust, Scar, and Sponge. The combination of these two models is investigated and discussed thoroughly in this paper. This work additionally employs a real industry dataset to confirm the feasibility of the proposed method and make sure it is applicable to real-world scenarios. The experimental results show that the proposed method can achieve higher precision and recall scores in steel surface defect detection.

AB - Steel is the most important material in the world of engineering and construction. Modern steelmaking relies on computer vision technologies, like optical cameras to monitor the production and manufacturing processes, which helps companies improve product quality. In this paper, we propose a deep learning method to automatically detect defects on the steel surface. The architecture of our proposed system is separated into two parts. The first part uses a revised version of single shot multibox detector (SSD) model to learn possible defects. Then, deep residual network (ResNet) is used to classify three types of defects: Rust, Scar, and Sponge. The combination of these two models is investigated and discussed thoroughly in this paper. This work additionally employs a real industry dataset to confirm the feasibility of the proposed method and make sure it is applicable to real-world scenarios. The experimental results show that the proposed method can achieve higher precision and recall scores in steel surface defect detection.

KW - Defect detection

KW - Fully convolutional networks

KW - ResNet

KW - SSD

UR - http://www.scopus.com/inward/record.url?scp=85067687403&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-20454-9_20

DO - 10.1007/978-3-030-20454-9_20

M3 - 會議論文篇章

AN - SCOPUS:85067687403

SN - 9783030204532

T3 - Advances in Intelligent Systems and Computing

SP - 202

EP - 212

BT - Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019

A2 - Ahram, Tareq

PB - Springer Verlag

T2 - AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019

Y2 - 24 July 2019 through 28 July 2019

ER -

Lin CY, Chen CH, Yang CY, Akhyar F, Hsu CY, Ng HF. Cascading Convolutional Neural Network for Steel Surface Defect Detection. In Ahram T, editor, Advances in Artificial Intelligence, Software and Systems Engineering - Proceedings of the AHFE International Conference on Human Factors in Artificial Intelligence and Social Computing, the AHFE International Conference on Human Factors, Software, Service and Systems Engineering, and the AHFE International Conference of Human Factors in Energy, 2019. Springer Verlag. 2020. p. 202-212. (Advances in Intelligent Systems and Computing). doi: 10.1007/978-3-030-20454-9_20

Cascading Convolutional Neural Network for Steel Surface Defect Detection

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

The Intelligent Parter of the Blind( II )

Cite this