Road Defect Detection System Based on Deep Learning
DOI:
https://doi.org/10.71222/n6p5hq54Keywords:
YOLOv8, deep learning, road defect, real-time detection, PyQt5Abstract
Road defect detection plays a vital role in the development of smart cities, enhancing the efficiency of road maintenance and providing reliable perception information for automated repair and inspection technologies. This article presents a deep learning-based road defect detection model, built upon advanced frameworks such as YOLOv8 and YOLOv5. Trained on a large-scale dataset of road images, the model is capable of accurately identifying common defect types, including cracks, potholes, and pavement deterioration. In addition, we developed a comprehensive road defect detection system featuring a user-friendly graphical interface, which supports real-time detection and visualization of road defects. Implemented using Python and PyQt5, the system enables intuitive display of detection results and provides detailed information for road maintenance planning. The proposed approach demonstrates promising performance in both static image recognition and continuous video monitoring, offering potential applications in intelligent transportation systems, urban road management, and automated infrastructure maintenance.
References
1. L. Zhang, F. Yang, Y. D. Zhang, Y. J. Zhu, "Road crack detection using deep convolutional neural network," 2016 IEEE Int. Conf. Image Process. (ICIP), IEEE, 2016, doi: 10.1109/ICIP.2016.7533052.
2. H. Maeda, et al., " Road damage detection and classification using deep learning," Mach. Intell. Cybern., vol. 4, no. 2, pp. 1–8, 2018, doi: 10.1111/mice.12387.
3. A. Alfarrarjeh, D. Trivedi, S. H. Kim, C. Shahabi, "A deep learning approach for road damage detection from smartphone images," 2018 IEEE Int. Conf. Big Data (Big Data), IEEE, 2018, doi: 10.1109/BigData.2018.8621899.
4. K. Gopalakrishnan, et al., "Deep convolutional neural networks with transfer learning for computer vision-based data-driven pavement distress detection," Constr. Build. Mater., vol. 157, pp. 322-330, 2017, doi: 10.1016/j.conbuildmat.2017.09.110.
5. M. Eisenbach, et al., "How to get pavement distress detection ready for deep learning? A systematic approach," 2017 Int. Joint Conf. Neural Netw. (IJCNN), IEEE, 2017, doi: 10.1109/IJCNN.2017.7966101.
6. E. Zalama, J. Gómez‐García‐Bermejo, R. Medina, and J. Llamas, "Road crack detection using visual features extracted by Gabor filters," Comput.-Aided Civ. Infrastruct. Eng., vol. 29, no. 5, pp. 342–358, 2014, doi: 10.1111/mice.12042.
7. S. Yang, “The Impact of Continuous Integration and Continuous Delivery on Software Development Efficiency”, J. Comput. Signal Syst. Res., vol. 2, no. 3, pp. 59–68, Apr. 2025, doi: 10.71222/pzvfqm21.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Shijie Lin (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
