Light & Engineering 34 (3) 2026

Volume 34 (3)
Pages 4-12

Purchase PDF - ₽600

Detection of Road Defects from Unmanned Aerial Vehicles Using Machine Learning L&E, Vol.34, No.3, 2026

Articles authors:

Ahsan Mustafa, Michael Yu. Kataev, Dmitry S. Efremenko

Ahsan Mustafa is a software engineer and researcher, working towards specializing in computer vision and neural networks, with a focus on UAV-based road defect detection. He holds a Bachelor's in Computer Science from the National University of Computer and Emerging Sciences (Islamabad, Pakistan) and is currently advancing his studies in Russia under the Open Doors Scholarship. His research field include AIdriven optimization and advanced imaging techniques

Michael Yu. Kataev, Doctor of Technical Science. He graduated in 1984 from Tomsk State University as an engineer-researcher in optic. At present, he is a Professor of the Department of Automated Control Systems (ACS) TU-RMS and Professor of Yurginsky Technological Institute, branch of the Tomsk Polytechnic University (TPU), scientific leadership, and a member of the Earth from Space Monitoring Centre, TUSUR

Dmitry S. Efremenko, Doctor of Technical Sciences, Corresponding Member of the Academy of Electro-technical Sciences, RF. He graduated from MPEI in 2009, Research scientist, Associate Professor. In 2017, he received the Elsevier/JQSRT Goody Award, he has more than 80 scientific papers in the field of radiation transfer theory

Abstract:

The timely detection of road surface deterioration is a critical factor in infrastructure maintenance and traffic safety. Conventional inspection methods remain labour-intensive, costly, and prone to human error. In contrast, unmanned aerial vehicles (UAVs) combined with advanced computer vision techniques provide a scalable and cost-effective alternative.
In this study, we propose a neural network model based on the YOLOv12 object detection framework for identifying road surface defects. The model is trained on a UAV-PDD2023 dataset containing over 2400 high-resolution aerial images annotated across six categories of road defects. Two model configurations with varying preprocessing steps and hyperparameters are evaluated to assess robustness and generalization.
The evaluation of the model efficiency was performed when RGB imagery is replaced with single-channel counterparts. Results show that the accuracy of defect detection remains largely unaffected: over 90 percent of the defects detected in RGB imagery were also correctly identified in single-channel images, with a performance variation of less than ±5 %. This indicates that a single trained model can be effectively applied to images with different spectral characteristics.

References:

1. Kataev, M. Yu., Dadonova, M.M., Efremenko, D.S. Illumination Correction of Multi-Time RGB Images Obtained with an Unmanned Aerial Vehicle // Light & Engineering, 2021, Vol. 29, # 2, pp. 50–58.
2. Kataev, M. Yu., Kartashov, E.Y., Avdeenko, V.D. Method for detecting road defects using images obtained from unmanned aerial vehicles // Computer Optics, 2023, Vol. 47, # 3.
3. Chen, S., Efremenko, D.S., Zhang, Z., Meng, L. In-Terrestrial Aquaculture Fields Mapping from High Resolution Remote Sensing Images //Light & Engineering, 2023, Vol. 31, # 5, pp. 135–142.
4. Ariff, M.F.M., Abidin, M.S.Z. UAV Photogrammetry for Road Defects Mapping // Journal of Advanced Geospatial Science and Technology, 2023, Vol. 3, # 1, pp. 1–14.
5. Zhang, Y., Zuo, Z., Xu, X., Wu, J., Zhu, J., Zhang, H., Wang, J., Tian, Y. Road damage detection using UAV images based on multi-level attention mechanism // Automation in Construction, 2022, 144:104613.
6. He, X., Tang, Z., Deng, Y., Zhou, G., Wang, Y., Li, L. UAV-based road crack object-detection algorithm // Automation in Construction, 2023, 154:105014.
7. Zhang, Y., Chen, J., Wu, Z., Guo, X., and Jia, S. Optimizing pavement distress detection with UAV: A comparative study of vision transformer and convolutional neural networks // KSCE Journal of Civil Engineering, 2025, 29 (6):100095.
8. Sun, Z., Zhu, L., Qin, S., Yu, Y., Ju, R. and Li, Q. Road Surface Defect Detection Algorithm Based on YOLOv8 // Electronics, 2024, 13 (12): 2413.
9. Wang, J., Meng, R., Huang, Y., Zhou, L., Huo, L., Qiao, Z., Niu, C. Road defect detection based on improved YOLOv8s model // Scientific Reports, 2024, 14.
10. Li, J., Yuan, C., Wang, X. Real-time instance-level detection of asphalt pavement distress combining space-to-depth (SPD) YOLO and omni-scale network (OSNet) // Automation in Construction, 2023, 155:105062.
11. Alzamzami, O., Babour, A., Baalawi, W., Al Khuzayem, L. PDS-UAV: A Deep Learning-Based Pothole Detection System Using Unmanned Aerial Vehicle Images // Sustainability, 2024, 16 (21): 9168.
12. Silva, L., Leithardt, V. Q., Batista, V., Villarrubia Gonzalez, G., De Paz Santana, J.F. Automated Road Damage Detection Using UAV Images and Deep Learning Techniques // IEEE Access, 2023, # 11, pp. 62918–62931.
13. Chen, Y., Wang, Y., Luo, H, Li, X., Zhan, J., Chen, W. BCGW-YOLO: A lightweight network for road damage detection using enhanced feature fusion and dynamically adjusted gradient loss // Digital Signal Processing, 2026, 168:105609.
14. Sharma, S., Gosain, A. Addressing class imbalance in remote sensing using deep learning approaches: a systematic literature review // Evolutionary Intelligence, 2025, 18 (1).
15. Yan, H. and Zhang, J. UAV-PDD2023: A benchmark dataset for pavement distress detection based on UAV images // Data in Brief, 2023, 51:109692.
16. Tian, Y., Ye, Q., Doermann, D. YOLOv12: Attention-Centric Real-Time Object Detectors // arXiv preprint arXiv:2502.12524, 2025.
17. Kostenetskiy, P. S., Chulkevich, R. A., Kozyrev, V. I. HPC Resources of the Higher School of Economics // Journal of Physics: Conference Series, 2021, 1740 (1): 012050. DOI 10.1088/1742–6596/1740/1/012050.

Keywords

DOI: https://doi.org/10.33383/2025-075

Article in RUS:
Svetotekhnika # 1, 2026, pp. 75–81

Buy