Research on tunnel defect detection based on YOLOv8 model

Xuefeng Zhao; Huajie Zhu; Hongxia Dong; Yanjun Hao

doi:10.1117/12.3054638

20 December 2024 Research on tunnel defect detection based on YOLOv8 model

Xuefeng Zhao, Huajie Zhu, Hongxia Dong, Yanjun Hao

Author Affiliations +

Proceedings Volume 13421, Eighth International Conference on Traffic Engineering and Transportation System (ICTETS 2024); 134210T (2024) https://doi.org/10.1117/12.3054638
Event: Eighth International Conference on Traffic Engineering and Transportation System (ICTETS 2024), 2024, Dalian, China

Abstract

To enhance tunnel maintenance and traffic safety, this study implements automatic detection and identification of tunnel defects using the YOLOv8 model. By collecting defect image data from various tunnel environments and performing data preprocessing and annotation, the YOLOv8 model was constructed for defect detection. The results demonstrate that the YOLOv8 model achieves a precision of 0.87, a recall of 0.89, and an mAP50 of 0.87 on the test set, indicating excellent detection capabilities and accuracy. The model accurately identifies and locates defect types such as cracks, damage, and spalling. Visual analysis of the model's detection results confirms its ability to precisely locate and recognize various internal tunnel defects, further validating the model's stability and practicality. This research provides significant technical support for intelligent transportation and tunnel safety management, highlighting the importance of early identification and effective treatment of tunnel defects. It contributes to improving the safety and efficiency of tunnel operations.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Xuefeng Zhao, Huajie Zhu, Hongxia Dong, and Yanjun Hao "Research on tunnel defect detection based on YOLOv8 model", Proc. SPIE 13421, Eighth International Conference on Traffic Engineering and Transportation System (ICTETS 2024), 134210T (20 December 2024); https://doi.org/10.1117/12.3054638

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE