Unsupervised Anomaly Detection for Surface Defects With Dual-Siamese Network

Tao, X; Zhang, D; Ma, W; Hou, Z; Lu, ZF; Adak, C

Unsupervised Anomaly Detection for Surface Defects With Dual-Siamese Network

Tao, X Zhang, D Ma, W Hou, Z Lu, ZF Adak, C

Permalink

Publisher:: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Industrial Informatics, 2022, 18, (11), pp. 7707-7717
Issue Date:: 2022-11-01

Closed Access

	Filename	Description	Size
	Unsupervised_Anomaly_Detection_for_Surface_Defects_With_Dual-Siamese_Network.pdf	Published version	3.92 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Tao, X
dc.contributor.author	Zhang, D
dc.contributor.author	Ma, W
dc.contributor.author	Hou, Z
dc.contributor.author	Lu, ZF
dc.contributor.author	Adak, C https://orcid.org/0000-0002-9085-2770
dc.date.accessioned	2022-12-07T20:55:33Z
dc.date.available	2022-12-07T20:55:33Z
dc.date.issued	2022-11-01
dc.identifier.citation	IEEE Transactions on Industrial Informatics, 2022, 18, (11), pp. 7707-7717
dc.identifier.issn	1551-3203
dc.identifier.issn	1941-0050
dc.identifier.uri	http://hdl.handle.net/10453/164224
dc.description.abstract	Unsupervised anomaly detection in real industrial scenarios is challenging since the small amount of defect-free images contain limited discriminative information, and anomaly defects are unpredictable. Although nowadays image reconstruction-based methods are widely being used in various anomaly detection applications, they cannot effectively learn semantic representation, which leads to imperfect reconstruction. In this article, anomaly detection is formulated as a joint problem of feature reconstruction and inpainting in the dual-siamese framework. The proposed approach forces the network to model the feature distribution from the normal area and capture the semantic context for discriminating normal and abnormal areas. It first uses a Siamese architecture to capture discriminative features of defect-free samples and its corresponding defective samples generated by the defect random generation module. A dense feature fusion module is then employed to obtain the dense feature representation of dual input. The second Siamese network is proposed to reconstruct and inpaint the dual-dense features of the previous stage. Compared to the existing methods that mostly employ single image reconstruction, it is beneficial to simultaneously reconstruct and inpaint the information of dense discriminative features. The experimental results on the MVTec AD datasets and some major real industrial datasets demonstrate that our method achieves state-of-the-art inspection accuracy.
dc.language	English
dc.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.relation.ispartof	IEEE Transactions on Industrial Informatics
dc.relation.isbasedon	10.1109/TII.2022.3142326
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	08 Information and Computing Sciences, 09 Engineering, 10 Technology
dc.subject.classification	Electrical & Electronic Engineering
dc.title	Unsupervised Anomaly Detection for Surface Defects With Dual-Siamese Network
dc.type	Journal Article
utslib.citation.volume	18
utslib.for	08 Information and Computing Sciences
utslib.for	09 Engineering
utslib.for	10 Technology
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access	*
dc.date.updated	2022-12-07T20:54:59Z
pubs.issue	11
pubs.publication-status	Published
pubs.volume	18
utslib.citation.issue	11

Abstract:

Unsupervised anomaly detection in real industrial scenarios is challenging since the small amount of defect-free images contain limited discriminative information, and anomaly defects are unpredictable. Although nowadays image reconstruction-based methods are widely being used in various anomaly detection applications, they cannot effectively learn semantic representation, which leads to imperfect reconstruction. In this article, anomaly detection is formulated as a joint problem of feature reconstruction and inpainting in the dual-siamese framework. The proposed approach forces the network to model the feature distribution from the normal area and capture the semantic context for discriminating normal and abnormal areas. It first uses a Siamese architecture to capture discriminative features of defect-free samples and its corresponding defective samples generated by the defect random generation module. A dense feature fusion module is then employed to obtain the dense feature representation of dual input. The second Siamese network is proposed to reconstruct and inpaint the dual-dense features of the previous stage. Compared to the existing methods that mostly employ single image reconstruction, it is beneficial to simultaneously reconstruct and inpaint the information of dense discriminative features. The experimental results on the MVTec AD datasets and some major real industrial datasets demonstrate that our method achieves state-of-the-art inspection accuracy.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/164224