Industrial anomaly detection and one-class classification using generative adversarial networks

Y. T.K. Lai; Jwu-Sheng Hu; Y. H. Tsai; W. Y. Chiu

doi:10.1109/AIM.2018.8452228

Industrial anomaly detection and one-class classification using generative adversarial networks

Y. T.K. Lai, Jwu-Sheng Hu, Y. H. Tsai, W. Y. Chiu

Institute of Electrical and Control Engineering

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

1 Scopus citations

Abstract

Industrial image datasets for quality inspection are mostly sparse in defects. It is then hard for both automated optical inspection (AOI) machines and simple neural network classifiers to inspect all defects effectively. In this work, we develop a novel framework for industrial anomaly detection in one-class classification manner, which utilized pre-trained generative adversarial networks (GANs) as the rule of thumb to perform anomaly detection. Our results show that GANs are able to capture arbitrary and structural industrial images and can effectively discern defects when the query images are defective.

Original language	English
Title of host publication	AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1444-1449
Number of pages	6
ISBN (Print)	9781538618547
DOIs	https://doi.org/10.1109/AIM.2018.8452228
State	Published - 30 Aug 2018
Event	2018 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2018 - Auckland, New Zealand Duration: 9 Jul 2018 → 12 Jul 2018

Publication series

Name	IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM
Volume	2018-July

Conference

Conference	2018 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2018
Country	New Zealand
City	Auckland
Period	9/07/18 → 12/07/18

Access to Document

10.1109/AIM.2018.8452228

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Industrial anomaly detection and one-class classification using generative adversarial networks'. Together they form a unique fingerprint.

Cite this

Lai, Y. T. K., Hu, J-S., Tsai, Y. H., & Chiu, W. Y. (2018). Industrial anomaly detection and one-class classification using generative adversarial networks. In AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics (pp. 1444-1449). [8452228] (IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM; Vol. 2018-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AIM.2018.8452228

Lai, Y. T.K. ; Hu, Jwu-Sheng ; Tsai, Y. H. ; Chiu, W. Y. / Industrial anomaly detection and one-class classification using generative adversarial networks. AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1444-1449 (IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM).

@inproceedings{a74b7173b786460d869a785b47eefddd,

title = "Industrial anomaly detection and one-class classification using generative adversarial networks",

abstract = "Industrial image datasets for quality inspection are mostly sparse in defects. It is then hard for both automated optical inspection (AOI) machines and simple neural network classifiers to inspect all defects effectively. In this work, we develop a novel framework for industrial anomaly detection in one-class classification manner, which utilized pre-trained generative adversarial networks (GANs) as the rule of thumb to perform anomaly detection. Our results show that GANs are able to capture arbitrary and structural industrial images and can effectively discern defects when the query images are defective.",

author = "Lai, {Y. T.K.} and Jwu-Sheng Hu and Tsai, {Y. H.} and Chiu, {W. Y.}",

year = "2018",

month = aug,

day = "30",

doi = "10.1109/AIM.2018.8452228",

language = "English",

isbn = "9781538618547",

series = "IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1444--1449",

booktitle = "AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics",

note = "null ; Conference date: 09-07-2018 Through 12-07-2018",

}

Lai, YTK, Hu, J-S, Tsai, YH & Chiu, WY 2018, Industrial anomaly detection and one-class classification using generative adversarial networks. in AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics., 8452228, IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM, vol. 2018-July, Institute of Electrical and Electronics Engineers Inc., pp. 1444-1449, 2018 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2018, Auckland, New Zealand, 9/07/18. https://doi.org/10.1109/AIM.2018.8452228

Industrial anomaly detection and one-class classification using generative adversarial networks. / Lai, Y. T.K.; Hu, Jwu-Sheng; Tsai, Y. H.; Chiu, W. Y.

AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1444-1449 8452228 (IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM; Vol. 2018-July).

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

TY - GEN

T1 - Industrial anomaly detection and one-class classification using generative adversarial networks

AU - Lai, Y. T.K.

AU - Hu, Jwu-Sheng

AU - Tsai, Y. H.

AU - Chiu, W. Y.

PY - 2018/8/30

Y1 - 2018/8/30

N2 - Industrial image datasets for quality inspection are mostly sparse in defects. It is then hard for both automated optical inspection (AOI) machines and simple neural network classifiers to inspect all defects effectively. In this work, we develop a novel framework for industrial anomaly detection in one-class classification manner, which utilized pre-trained generative adversarial networks (GANs) as the rule of thumb to perform anomaly detection. Our results show that GANs are able to capture arbitrary and structural industrial images and can effectively discern defects when the query images are defective.

AB - Industrial image datasets for quality inspection are mostly sparse in defects. It is then hard for both automated optical inspection (AOI) machines and simple neural network classifiers to inspect all defects effectively. In this work, we develop a novel framework for industrial anomaly detection in one-class classification manner, which utilized pre-trained generative adversarial networks (GANs) as the rule of thumb to perform anomaly detection. Our results show that GANs are able to capture arbitrary and structural industrial images and can effectively discern defects when the query images are defective.

UR - http://www.scopus.com/inward/record.url?scp=85053909141&partnerID=8YFLogxK

U2 - 10.1109/AIM.2018.8452228

DO - 10.1109/AIM.2018.8452228

M3 - Conference contribution

AN - SCOPUS:85053909141

SN - 9781538618547

T3 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM

SP - 1444

EP - 1449

BT - AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 9 July 2018 through 12 July 2018

ER -

Lai YTK, Hu J-S, Tsai YH, Chiu WY. Industrial anomaly detection and one-class classification using generative adversarial networks. In AIM 2018 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1444-1449. 8452228. (IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM). https://doi.org/10.1109/AIM.2018.8452228