Computational Technology Resources - CCC

Y. Yang¹, W. Liu¹, J. Liu¹ and T. Xin²

¹Department of Engineering Mechanics, KTH Royal Institute of Technology, Stockholm, Sweden
²Standard Metrology Institute, China Academy of Railway Sciences, Beijing, China

doi:10.4203/ccc.15.10.12

Y. Yang, W. Liu, J. Liu, T. Xin, "Image-Based Crack Analysis of Railway Contact Wires Using a Two-Stage U-Net", in J. Pombo, (Editor), "Proceedings of the Seventh International Conference on Railway Technology: Research, Development and Maintenance ", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 15, Paper 10.12, 2026, doi:10.4203/ccc.15.10.12

Keywords: railway catenary wire, keyframe extraction, crack segmentation, skeletonization, deep learning, geometric measurement.

Abstract

The railway catenary system is essential for continuous electric power supply, and regular inspection is required to maintain safe and reliable long-term operation. As the component in direct contact with the pantograph, the contact wire is exposed to repeated mechanical and electrical loading, making surface cracking an important condition indicator. However, contact wire videos from operating railways are difficult to obtain. This paper therefore uses laboratory cyclic-loading videos as a controlled crack observation platform. We extract maximum-stroke keyframes to construct standardised contact wire keyframes for crack analysis. Standardised contact wire image regions are then segmented using a two-stage U-Net model, followed by mask cleaning, skeletonisation and pixel-level geometric measurement. The workflow provides measurable crack information for a safety-critical current-collection component, offering a basis for future real world contact wire inspection and maintenance decision support.

download the full-text of this paper (PDF, 10 pages, 988 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 15 PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE Edited by: J. Pombo Paper 10.12 Image-Based Crack Analysis of Railway Contact Wires Using a Two-Stage U-Net Y. Yang¹, W. Liu¹, J. Liu¹ and T. Xin² ¹Department of Engineering Mechanics, KTH Royal Institute of Technology, Stockholm, Sweden ²Standard Metrology Institute, China Academy of Railway Sciences, Beijing, China doi:10.4203/ccc.15.10.12 Full Bibliographic Reference for this paper Y. Yang, W. Liu, J. Liu, T. Xin, "Image-Based Crack Analysis of Railway Contact Wires Using a Two-Stage U-Net", in J. Pombo, (Editor), "Proceedings of the Seventh International Conference on Railway Technology: Research, Development and Maintenance ", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 15, Paper 10.12, 2026, doi:10.4203/ccc.15.10.12 Keywords: railway catenary wire, keyframe extraction, crack segmentation, skeletonization, deep learning, geometric measurement. Abstract The railway catenary system is essential for continuous electric power supply, and regular inspection is required to maintain safe and reliable long-term operation. As the component in direct contact with the pantograph, the contact wire is exposed to repeated mechanical and electrical loading, making surface cracking an important condition indicator. However, contact wire videos from operating railways are difficult to obtain. This paper therefore uses laboratory cyclic-loading videos as a controlled crack observation platform. We extract maximum-stroke keyframes to construct standardised contact wire keyframes for crack analysis. Standardised contact wire image regions are then segmented using a two-stage U-Net model, followed by mask cleaning, skeletonisation and pixel-level geometric measurement. The workflow provides measurable crack information for a safety-critical current-collection component, offering a basis for future real world contact wire inspection and maintenance decision support. download the full-text of this paper (PDF, 10 pages, 988 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions