Machine Learning and Computer Vision Applications in Civil Infrastructure Inspection and Monitoring

Shuming, L; Andy, G; Liang, B; Zhidong, L; Yu, D; Yang, W; Fang, C

Machine Learning and Computer Vision Applications in Civil Infrastructure Inspection and Monitoring

Shuming, L Andy, G Liang, B

Zhidong, L Yu, D Yang, W Fang, C

Permalink

Publisher:: John Wiley & Sons
Publication Type:: Chapter
Citation:: Infrastructure Robotics: Methodologies, Robotic Systems and Applications, 2023, pp. 59-80
Issue Date:: 2023-12-13

Closed Access

	Filename	Description	Size
	Chapter_4.pdf	Published version	2.53 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	Shuming, L
dc.contributor.author	Andy, G
dc.contributor.author	Liang, B https://orcid.org/0000-0002-6605-2167
dc.contributor.author	Zhidong, L
dc.contributor.author	Yu, D
dc.contributor.author	Yang, W
dc.contributor.author	Fang, C
dc.contributor.editor	Dikai, L
dc.contributor.editor	Carlos, B
dc.contributor.editor	Gamini, D
dc.contributor.editor	Mirko, K
dc.date.accessioned	2024-01-15T21:37:46Z
dc.date.available	2024-01-15T21:37:46Z
dc.date.issued	2023-12-13
dc.identifier.citation	Infrastructure Robotics: Methodologies, Robotic Systems and Applications, 2023, pp. 59-80
dc.identifier.isbn	1394162863
dc.identifier.isbn	9781394162864
dc.identifier.uri	http://hdl.handle.net/10453/174503
dc.description.abstract	Civil infrastructure is one of the main drivers that enable economic growth and ensure a high standard of living in any country. Therefore, ensuring the safety and serviceability of existing civil infrastructure is of paramount importance to support vital economic activities and to prevent any sudden infrastructure failure that may engender catastrophic consequences. With the advancements in machine learning and computer vision techniques, recent years have seen the emergence of advanced data-driven solutions to address the challenges in the areas of civil infrastructure inspection and monitoring. This chapter covers the details of two successful applications using machine learning and computer vision in water utilities and transport. The pipe failure prediction framework based on graph neural networks ( GNN ) jointly considers the pipes' features, the network structure, the geographical neighboring effect, and the temporal failure series, achieving state-of-the-art prediction performance. This framework provides the water utility with core data-driven support for proactive maintenance such as regular pipe inspection, pipe renewal planning, and sensor system deployment. Furthermore, a novel two-stage detection model is proposed to effectively capture train signal aspect transitions. The model leverages state-of-the-art computer vision algorithms and is trained on video footage data. The experimental results showcase the model's high accuracy in detecting and predicting signal aspect states and their transitions.
dc.language	en
dc.publisher	John Wiley & Sons
dc.relation.ispartof	Infrastructure Robotics: Methodologies, Robotic Systems and Applications
dc.relation.isbasedon	10.1002/9781394162871.ch4
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Machine Learning and Computer Vision Applications in Civil Infrastructure Inspection and Monitoring
dc.type	Chapter
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/A/DRsch The Data Science Institute
utslib.copyright.status	closed_access	*
pubs.consider-herdc	true
dc.date.updated	2024-01-15T21:37:45Z
pubs.publication-status	Published

Abstract:

Civil infrastructure is one of the main drivers that enable economic growth and ensure a high standard of living in any country. Therefore, ensuring the safety and serviceability of existing civil infrastructure is of paramount importance to support vital economic activities and to prevent any sudden infrastructure failure that may engender catastrophic consequences. With the advancements in machine learning and computer vision techniques, recent years have seen the emergence of advanced data-driven solutions to address the challenges in the areas of civil infrastructure inspection and monitoring. This chapter covers the details of two successful applications using machine learning and computer vision in water utilities and transport. The pipe failure prediction framework based on graph neural networks ( GNN ) jointly considers the pipes' features, the network structure, the geographical neighboring effect, and the temporal failure series, achieving state-of-the-art prediction performance. This framework provides the water utility with core data-driven support for proactive maintenance such as regular pipe inspection, pipe renewal planning, and sensor system deployment. Furthermore, a novel two-stage detection model is proposed to effectively capture train signal aspect transitions. The model leverages state-of-the-art computer vision algorithms and is trained on video footage data. The experimental results showcase the model's high accuracy in detecting and predicting signal aspect states and their transitions.

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

http://hdl.handle.net/10453/174503