Convolutional neural network for detecting railway fastener defects using a developed 3D laser system

You Zhan, Xianxing Dai, Enhui Yang and Kelvin C.P. Wang

International Journal of Rail Transportation, 2021, vol. 9, issue 5, 424-444

Abstract: This study develops a three-dimensional (3D) Laser Railway Detection System for automated railway fastener defect detection on 3D ballastless track. The 3D laser imaging system overcomes the shortcomings of shadows and illumination variations, thereby providing 3D information of the ballastless track with high reproducibility and accuracy. RailNet, an efficient architecture based on a Convolutional Neural Network (CNN), is proposed in this paper for detecting high-speed railway fastener defects on 3D ballastless track. RailNet consists of 10 layers and includes more than 120,000 parameters. RailNet is trained using 80,000 3D fastener images with 1-mm resolution and is then demonstrated to be successful at identifying damaged and missing fasteners. The testing results show that the system described in this paper can inspect the defective hook-shaped fasteners notably well. The proposed RailNet significantly outperforms the other approaches with a prediction accuracy of 100%, and the number of testing samples is 16,000.

Date: 2021
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/23248378.2020.1825128 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:tjrtxx:v:9:y:2021:i:5:p:424-444

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/tjrt20

DOI: 10.1080/23248378.2020.1825128

Access Statistics for this article

International Journal of Rail Transportation is currently edited by Wanming Zhai and Kelvin C. P. Wang

More articles in International Journal of Rail Transportation from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().