Detection method of surface damage of concrete bridge based on improved YOLOv8

Yongrui Zhang; Wei Wu; Jiaqi Huang; Lee Cong

doi:10.1117/12.3029326

6 May 2024 Detection method of surface damage of concrete bridge based on improved YOLOv8

Yongrui Zhang, Wei Wu, Jiaqi Huang, Lee Cong

Proceedings Volume 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024); 131071K (2024) https://doi.org/10.1117/12.3029326
Event: Fourth International Conference on Sensors and Information Technology (ICSI 2024), 2024, Xiamen, China

Abstract

In the existing target detection algorithms, it has not been improved according to the characteristics of bridge surface damages, and the detection accuracy of bridge apparent diseases under complex background is low. To enhance the accuracy of concrete bridge surface damage detection in complex backgrounds, a bridge surface damage detection method based on the improved YOLOv8 algorithm is proposed. Firstly, addressing the characteristics of densely distributed damages and significant variations in damages scales, the network structure of YOLOv8 is modified by embedding the CBAM (Convolutional Block Attention Module) attention module into the detection layer. Experimental results demonstrate that the improved YOLOv8 model exhibits significant improvements in precision, recall, average classification accuracy, and other metrics compared to the original model. The overall mean average precision increased by 1.23%, indicating a more precise and real-time detection of bridge damages.

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

Citation Download Citation

Yongrui Zhang, Wei Wu, Jiaqi Huang, and Lee Cong "Detection method of surface damage of concrete bridge based on improved YOLOv8", Proc. SPIE 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024), 131071K (6 May 2024); https://doi.org/10.1117/12.3029326

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

Includes PDF, HTML & Video, when available