Deep Learning-Based Road Surface Defect Detection Using Image Processing Techniques

Shivangkumar R Patel; Karan P Bhatt; Jwalant Babubhai Baria; Vanrajkumar Dineshkumar Baria; Ashok K Rathva; Yagnik A Rathod

doi:10.29284/a9q1q517

Authors

Shivangkumar R Patel Assistant Professor, Computer Engineering Department, Government Engineering College, Modasa
Karan P Bhatt Assistant Professor, Computer Engineering Department, Vishwakarma Government Engineering College, Chandkheda
Jwalant Babubhai Baria Assistant Professor, Computer Engineering Department, Government Engineering College, Dahod
Vanrajkumar Dineshkumar Baria Assistant Professor, Computer Engineering Department, Government Engineering College, Dahod
Ashok K Rathva Assistant Professor, Computer Engineering Department, Government Engineering College, Dahod
Yagnik A Rathod Assistant Professor, Computer Engineering Department, Government Engineering College, Dahod

DOI:

https://doi.org/10.29284/a9q1q517

Keywords:

Road Surface Defect Detection, Deep Learning, Pavement Crack Detection, Object Detection Networks, Image Processing Techniques, YOLO-based Detection, Smart Infrastructure Monitoring.

Abstract

The importance of detecting road surface defects (cracks and potholes) promptly is crucial to transportation safety, minimisation of maintenance plans and minimisation of infrastructure deterioration costs. Conventional inspection techniques are time-consuming, subjective, and can hardly be used to offer scalable solutions to large road networks. The recent developments in deep learning and computer vision have shown that automated pavement distress detection is highly promising. The proposed research is a deep learning framework of road surface defect detection that combines image preprocessing with the lightweight object detection networks. The suggested methodology will also use contrast enhancement, noise reduction, and data augmentation to enhance the model's strength when faced with different illuminations and environmental conditions. Real-time defect localisation and classification make use of a convolutional neural network-based detection architecture, which is based on the state-of-the-art object detection models. The evaluation of the performance is based on typical measures such as precision, recall, F1-score, and the mean Average Precision (mAP). The superiority in accuracy of detection and the computational efficiency is proven by the comparisons with the already developed deep learning methods. The findings show that the combination of image enhancement methods and optimised deep neural networks can help a great deal to achieve detection reliability in real-world scenarios. The suggested architecture offers a highly scalable and low-cost solution to an intelligent pavement monitoring system and smart infrastructure management.

References

1. Arya, D., Maeda, H., Ghosh, S. K., Toshniwal, D., & Mraz, A. (2021). Global road damage detection using deep learning. IEEE Access, 9, 160162–160176. https://doi.org/10.1109/ACCESS.2021.3132203

2. Bochkovskiy, A., Wang, C. Y., & Liao, H. Y. M. (2020). YOLOv4: Optimal speed and accuracy of object detection. arXiv. https://doi.org/10.48550/arXiv.2004.10934

3. Canny, J. (1986). A computational approach to edge detection. IEEE TPAMI, 8(6), 679–698. https://doi.org/10.1109/TPAMI.1986.4767851

4. Cha, Y. J., Choi, W., & Büyüköztürk, O. (2017). Deep learning-based crack damage detection using convolutional neural networks. Computer-Aided Civil and Infrastructure Engineering, 32(5), 361–378. https://doi.org/10.1111/mice.12263

5. Chen, L. C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. (2018). DeepLabv3+: Encoder-decoder with atrous convolution. ECCV. https://doi.org/10.1007/978-3-030-01234-2_49

6. Chen, X., Yongchareon, S., & Knoche, M. (2023). A review on computer vision and machine learning techniques for automated road surface defect and distress detection. Journal of Smart Cities and Society, 1(4), 259-275.

7. Cumbajin, E., Rodrigues, N., Costa, P., Miragaia, R., Frazão, L., Costa, N., ... & Pereira, A. (2023). A systematic review on deep learning with CNNs applied to surface defect detection. Journal of Imaging, 9(10), 193.

8. Eisenbach, M., Stricker, R., Seichter, D., Amende, K., Debes, K., Sesselmann, M., & Gross, H. M. (2017). How to get pavement distress detection ready for deep learning? International Joint Conference on Neural Networks (IJCNN). https://doi.org/10.1109/IJCNN.2017.7966211

9. Fan, L., Wang, D., Wang, J., Li, Y., Cao, Y., Liu, Y., ... & Wang, Y. (2023). Pavement defect detection with deep learning: A comprehensive survey. IEEE Transactions on Intelligent Vehicles, 9(3), 4292-4311.

10. Fan, R., Ai, M., & Li, Z. (2019). Road crack detection using a deep convolutional neural network and adaptive thresholding. IEEE Access, 7, 12384–12396. https://doi.org/10.1109/ACCESS.2019.2891167

11. Gopalakrishnan, K. (2018). Deep learning in data-driven pavement distress detection and classification. Transportation Research Record, 2672(45), 91–100. https://doi.org/10.1177/0361198118758059

12. Guo, Y., Yang, X., Li, H., & Liang, Y. (2020). Pavement crack detection using deep CNN and transfer learning. Sensors, 20(5), 1444. https://doi.org/10.3390/s20051444

13. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. CVPR. https://doi.org/10.1109/CVPR.2016.90

14. Huyan, J., Li, W., Tighe, S., Xu, H., Zhai, J., & Chen, Y. (2020). CrackU-Net: A novel deep convolutional neural network for pavement crack detection. Automation in Construction, 115. https://doi.org/10.1016/j.autcon.2020.103212

15. Li, P., Zhou, B., Wang, C., Hu, G., Yan, Y., Guo, R., & Xia, H. (2024). CNN-based pavement defects detection using grey and depth images. Automation in Construction, 158, 105192.

16. Li, S., Zhao, X., & Zhou, G. (2020). Automatic pixel-level pavement crack detection using information of multi-scale neighbourhoods. IEEE Access, 8, 60604–60614. https://doi.org/10.1109/ACCESS.2020.2983626

17. Lin, S. (2025). Road Defect Detection System Based on Deep Learning. Journal of Computer, Signal, and System Research, 2(5), 79-87.

18. Lin, T. Y., et al. (2014). Microsoft COCO: Common objects in context. ECCV. https://doi.org/10.1007/978-3-319-10602-1_48

19. Liu, K., Zheng, H., & He, X. (2025, February). Deep-learning-based road defect detection using an improved YOLOv8 model. In International Workshop on Advanced Imaging Technology (IWAIT) 2025 (Vol. 13510, pp. 139-145). SPIE.

20. Liu, Y., Zhang, C., & Dong, X. (2023). A survey of real-time surface defect inspection methods based on deep learning. Artificial Intelligence Review, 56(10), 12131-12170.

21. Liu, Z., Cao, Y., Wang, Y., & Wang, W. (2019). Computer vision-based concrete crack detection using U-Net fully convolutional networks. Automation in Construction, 104, 129–139. https://doi.org/10.1016/j.autcon.2019.04.005

22. Maeda, H., Sekimoto, Y., Seto, T., Kashiyama, T., & Omata, H. (2018). Road damage detection using deep neural networks with images captured through a smartphone. Computer-Aided Civil and Infrastructure Engineering, 33(12), 1127–1141. https://doi.org/10.1111/mice.12387

23. Maeda, H., Sekimoto, Y., Seto, T., Kashiyama, T., & Omata, H. (2020). Road damage detection and classification using deep neural networks with smartphone images. Computer-Aided Civil and Infrastructure Engineering, 35(10), 1045–1063*. https://doi.org/10.1111/mice.12547

24. Mir, T. A., Banerjee, D., Upadhyay, D., & Rawat, R. S. (2024, November). Smart Solutions for Road Defect Detection: A Deep Learning Fusion Approach with CNNs and SVMs. In 2024 4th International Conference on Technological Advancements in Computational Sciences (ICTACS) (pp. 230-235). IEEE.

25. Oliveira, H., & Correia, P. L. (2013). Automatic road crack detection and Characterisation. IEEE Transactions on Intelligent Transportation Systems, 14(1), 155–168. https://doi.org/10.1109/TITS.2012.2208630

26. Omarov, B., & Kulambayev, B. (2024). Deep Learning-Based Image Processing for Real-Time Detection of Road Surface Damage. Procedia Computer Science, 251, 609-614.

27. Otsu, N. (1979). A threshold selection method from grey-level histograms. IEEE Transactions on Systems, Man, and Cybernetics, 9(1), 62–66. https://doi.org/10.1109/TSMC.1979.4310076

28. Prunella, M., Scardigno, R. M., Buongiorno, D., Brunetti, A., Longo, N., Carli, R., ... & Bevilacqua, V. (2023). Deep learning for automatic vision-based recognition of industrial surface defects: A survey. IEEE Access, 11, 43370-43423.

29. Redmon, J., & Farhadi, A. (2018). YOLOv3: An incremental improvement. arXiv. https://doi.org/10.48550/arXiv.1804.02767

30. Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster R-CNN: Towards real-time object detection. IEEE TPAMI, 39(6), 1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031

31. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. MICCAI. https://doi.org/10.1007/978-3-319-24574-4_28

32. Sattar, K. A., Abdel-Nasser, M., El Ferik, S., & Taha, A. E. (2025). Detecting Road Defects and Hazards in Metropolitan Environments Using Optimized Deep Learning Techniques. Transportation Research Procedia, 84, 528-533.

33. Shorten, C., & Khoshgoftaar, T. (2019). A survey on image data augmentation for deep learning. Journal of Big Data, 6, 60. https://doi.org/10.1186/s40537-019-0197-0

34. Simonyan, K., & Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. https://doi.org/10.48550/arXiv.1409.1556

35. Ultralytics. (2023). YOLOv8: A state-of-the-art object detection model. GitHub Repository. (Used for methodological citation; no DOI but widely accepted reference)

36. Wang, Y., Yu, C., & Li, C. (2020). Road crack detection based on an improved U-Net. Sensors, 20(17), 4911. https://doi.org/10.3390/s20174911

37. Yang, X., Li, H., Yu, Y., Luo, X., Huang, T., & Yang, X. (2018). Automatic pixel-level crack detection and measurement using a fully convolutional network. Computer-Aided Civil and Infrastructure Engineering, 33(12), 1090–1109. https://doi.org/10.1111/mice.12393

38. Yu, J., Jiang, J., Fichera, S., Paoletti, P., Layzell, L., Mehta, D., & Luo, S. (2024). Road surface defect detection—from image-based to non-image-based: a survey. IEEE transactions on intelligent transportation Systems, 25(9), 10581-10603.

39. Yu, J., Jiang, J., Fichera, S., Paoletti, P., Layzell, L., Mehta, D., & Luo, S. (2024). Road surface defect detection—from image-based to non-image-based: a survey. IEEE transactions on intelligent transportation Systems, 25(9), 10581-10603.

40. Zhang, A., Wang, K. C. P., Li, B., Yang, E., Dai, X., Peng, Y., & Fei, Y. (2017). Automated pixel-level pavement crack detection on 3D asphalt surfaces using a deep-learning network. Computer-Aided Civil and Infrastructure Engineering, 32(10), 805–819. https://doi.org/10.1111/mice.12297

41. Zhang, L., Yang, F., Zhang, Y. D., & Zhu, Y. J. (2016). Road crack detection using a deep convolutional neural network. IEEE International Conference on Image Processing (ICIP). https://doi.org/10.1109/ICIP.2016.7533052

42. Zhang, W., Zhang, Z., Qi, D., & Liu, Y. (2021). Automatic crack detection and classification using deep learning. IEEE Access, 9, 21404–21413. https://doi.org/10.1109/ACCESS.2021.3055176

43. Zhang, Y., & Cheng, H. (2019). Automated pavement crack detection using deep learning. Automation in Construction, 104, 1–11. https://doi.org/10.1016/j.autcon.2019.03.015

44. Zhao, J., Zhang, C., Cong, S., Yu, Y., Yue, X., Shen, Y., & Hui, Y. (2025). Research on Road Defect Detection Based on Deep Learning. Engineering Letters, 33(9).