Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Advances in Computer, Signals and Systems
  4. Vol 10, Issue 1, 2026

Research on pavement disease detection algorithm based on YOLOv5s

Download as PDF

DOI: 10.23977/acss.2026.100107 | Downloads: 1 | Views: 56

Author(s)

Hanpeng Wu 1, Yuxi Guo 1, Zhengbing Zheng 1

Affiliation(s)

1 Shaanxi University of Technology, Hanzhong, Shaanxi, 723001, China

Corresponding Author

Hanpeng Wu

ABSTRACT

In response to the problems such as significant background interference, large differences in target scales, and low detection accuracy for small targets in the current pavement defect detection task, this paper proposes an improved YOLOv5s pavement defect detection algorithm. The training dataset is obtained by integrating the open-source RDD2022 dataset and the self-built dataset, and then the data is augmented through Mosaic method before being input into the network for training. In addition, the CBAM attention mechanism is introduced into the backbone network to enhance the model's ability to focus on the pavement defect features through both channel and spatial attention. Finally, the NWD optimization loss function is adopted to improve the model's detection accuracy for small targets and ablation experiments are conducted. The experimental results show that the improved algorithm has an [email protected] of 3.8% higher than that of the original YOLOv5s on the dataset, effectively enhancing the model's detection accuracy.

KEYWORDS

Pavement defect detection; data enhancement; CBAM; NWD

CITE THIS PAPER

Hanpeng Wu, Yuxi Guo, Zhengbing Zheng. Research on pavement disease detection algorithm based on YOLOv5s. Advances in Computer, Signals and Systems (2026). Vol. 10, No. 1, 54-61. DOI: http://dx.doi.org/10.23977/acss.2026.100107.

REFERENCES

[1] Ministry of Transport of the People's Republic of China. Statistical Bulletin on the Development of the Transportation Industry in 2024 [EB/OL]. The Central People's Government of the People's Republic of China website, 2025. https://xxgk.mot.gov.cn/2020/jigou/zhghs/202506/t20250610_4170228.html.
[2] Chen Yun. Research on the fusion of traditional and intelligent technology of highway pavement disease detection [J]. Intelligent Building and Smart City2025,(04):27-29.
[3] GIRSHICK Ross, DONAHUE Jeff, DARRELL Trevor, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]. IEEE Conference on Computer Vision and Pattern Recognition(CVPR). IEEE,Ohio,USA,2014:580-587
[4] Liu P Y, Yuan J, Gao Q, et al. Road surface disease detection method based on improved YOLOv5 [J]. Journal of Beijing University of Technology, 2025, 51 (05): 552-559.
[5] WOO Sanghyun, PARK Jongchan, LEE Joon Young, et al. CBAM: Convolutional Block Attention Module[C]. Spring:Cham,Switzerland,2018:3-19
[6] Jinwang Wang, Chang Xu, Wen Yang, Lei Yu. A Normalized Gaussian Wasserstein Distance for Tiny Object Detection[J]. arXiv preprint, arXiv; 2110.13389.2021.
[7] DEEKSHA Arya, HIROYA Maeda, SANJAY Kumar Ghosh, et al. RDD2022:A multi-national image dataset for automatic Road Damage Detection[J]. IEEEBigData,2022,1-22.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 43997
Visits: 969771

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.