ViTALnet: Anomaly on Industrial Textured Surfaces With Hybrid Transformer

Tao, X; Adak, C; Chun, PJ; Yan, S; Liu, H

ViTALnet: Anomaly on Industrial Textured Surfaces With Hybrid Transformer

Tao, X Adak, C

Chun, PJ Yan, S Liu, H

Permalink

Publisher:: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Instrumentation and Measurement, 2023, 72
Issue Date:: 2023-01-01

Closed Access

	Filename	Description	Size
	ViTALnet_Anomaly_on_Industrial_Textured_Surfaces_With_Hybrid_Transformer.pdf	Published version	19.99 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	Adak, C https://orcid.org/0000-0002-9085-2770
dc.contributor.author	Chun, PJ
dc.contributor.author	Yan, S
dc.contributor.author	Liu, H
dc.date.accessioned	2024-03-15T01:17:10Z
dc.date.available	2024-03-15T01:17:10Z
dc.date.issued	2023-01-01
dc.identifier.citation	IEEE Transactions on Instrumentation and Measurement, 2023, 72
dc.identifier.issn	0018-9456
dc.identifier.issn	1557-9662
dc.identifier.uri	http://hdl.handle.net/10453/176746
dc.description.abstract	The coexistence of subtle and long-range anomalies in real-world industrial applications brings significant challenges for anomaly localization. Existing methods typically train deep models by utilizing the multilevel patches or layers-fusion approaches for learning the global-local distribution; however, these methods do not consider learning local and global features simultaneously, which suffer from inaccurate localization results. To this end, a hybrid transformer model, ViTALnet, is proposed here, which is established based on fine-grained feature reconstruction. Our ViTALnet first adopts the vision transformer (ViT) to extract local discriminatory features as feature representation, which leverages the global semantic capturing capability. Then, an anomaly estimation module is proposed by integrating global attention and a pyramidal architecture to enhance contextual information for fine-grain anomaly localization. The experiments were extensively conducted on industrial anomaly localization datasets MVTec Anomaly Detection (MVTec AD)-Texture, NanoTWICE, and general textured datasets KolektorSDD2, MT Defect, and Dot-patterned Fabric, where our proposed ViTALnet outperformed major state-of-the-art (SOTA) methods.
dc.language	English
dc.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.relation.ispartof	IEEE Transactions on Instrumentation and Measurement
dc.relation.isbasedon	10.1109/TIM.2023.3250225
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0299 Other Physical Sciences, 0906 Electrical and Electronic Engineering
dc.subject.classification	Electrical & Electronic Engineering
dc.subject.classification	4006 Communications engineering
dc.subject.classification	4008 Electrical engineering
dc.subject.classification	4009 Electronics, sensors and digital hardware
dc.title	ViTALnet: Anomaly on Industrial Textured Surfaces With Hybrid Transformer
dc.type	Journal Article
utslib.citation.volume	72
utslib.for	0299 Other Physical Sciences
utslib.for	0906 Electrical and Electronic Engineering
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
dc.date.updated	2024-03-15T01:17:00Z
pubs.publication-status	Published
pubs.volume	72

Abstract:

The coexistence of subtle and long-range anomalies in real-world industrial applications brings significant challenges for anomaly localization. Existing methods typically train deep models by utilizing the multilevel patches or layers-fusion approaches for learning the global-local distribution; however, these methods do not consider learning local and global features simultaneously, which suffer from inaccurate localization results. To this end, a hybrid transformer model, ViTALnet, is proposed here, which is established based on fine-grained feature reconstruction. Our ViTALnet first adopts the vision transformer (ViT) to extract local discriminatory features as feature representation, which leverages the global semantic capturing capability. Then, an anomaly estimation module is proposed by integrating global attention and a pyramidal architecture to enhance contextual information for fine-grain anomaly localization. The experiments were extensively conducted on industrial anomaly localization datasets MVTec Anomaly Detection (MVTec AD)-Texture, NanoTWICE, and general textured datasets KolektorSDD2, MT Defect, and Dot-patterned Fabric, where our proposed ViTALnet outperformed major state-of-the-art (SOTA) methods.

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

http://hdl.handle.net/10453/176746