Non-Contact Bolt Detection Based on YOLOv5-Ganomaly Algorithm and UAV

DOI

10.2991/978-94-6463-398-6_80How to use a DOI?

Keywords

YOLOv5; Ganomaly; High-strength bolt; Anomaly detection; Non-contact detection; UAV

Abstract

High-strength bolt loosening detection is an important part of steel bridge inspection. The automatic detection methods based on machine vision and UAV have the characteristics of fast speed and high efficiency, and have been widely used to replace manual inspection. However, it is difficult to realize the automatic inspection based on vision due to the high height of the position where high-strength bolts are located, the small target, the loosening characteristics are not obvious, and the large number of characteristics. This paper proposes non-contact bolt detection to get the bolt image information, and then combine YOLOv5 and Ganomaly algorithm, propose YOLOv5-Ganomaly semi-supervised learning model bolt detection algorithm. Firstly, locate the bolt target area through YOLOv5-CT model, and automatically screen the un-lost bolts; then pre-process the un-lost bolt images, and finally detect the anomaly looseness of the pre-processed images through Ganomaly algorithm, and automatically determine the looseness of the bolts by the given threshold value. The test results show that the detection accuracy of lost bolt loosening reaches 98.3% and the accuracy of bolt loosening detection can reach 85%.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Haibo Xie
AU  - Qianyu Liao
AU  - ShiXiang Yang
AU  - Junwei Zhu
PY  - 2024
DA  - 2024/04/24
TI  - Non-Contact Bolt Detection Based on YOLOv5-Ganomaly Algorithm and UAV
BT  - Proceedings of the 2023 5th International Conference on Hydraulic, Civil and Construction Engineering (HCCE 2023)
PB  - Atlantis Press
SP  - 848
EP  - 857
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-398-6_80
DO  - 10.2991/978-94-6463-398-6_80
ID  - Xie2024
ER  -