Pothole Detection and Priority Ranking Using Deep Learning

  • Harsh Gupta Department of Computing Technologies SRM Institute of Science and Technology Chennai, India
  • P Rajeev Siddarth Department of Computing Technologies SRM Institute of Science and Technology Chennai, India
  • Dr P. S. Thanigaivelu Department of Computing Technologies SRM Institute of Science and Technology Chennai, India
DOI: https://doi.org/10.69980/cse.v12i1.6649
Keywords: Potholes, YOLOv8, Faster R-CNN, RetinaNet, deep learning-based framework

Abstract

Potholes on the road pose a great danger to transport safety and infrastructure performance. This paper presents a deep learning-based framework for pothole detection and priority ranking. Three object detection models are implemented and compared with each other in terms of precision, recall, and mAP: YOLOv8, Faster R-CNN, and RetinaNet. Experimental results show that YOLOv8 achieves the best overall performance with a balanced trade-off between accuracy and efficiency. Identified potholes are further classified into severity levels and ready to be incorporated with a context-aware prioritisation system. The suggested framework demonstrates a realistic implementation in the context of intelligent road maintenance, and the future research will be aimed at the optimisation-based ranking and practical implementation.

Downloads

Download data is not yet available.

References

J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016.

[2] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv preprint arXiv:2004.10934, 2020.

[3] G. Jocher et al., “Ultralytics YOLOv8,” 2023. [Online]. Available: https://github.com/ultralytics/ultralytics

[4] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137–1149, 2017.

[5] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal Loss for Dense Object Detection,” in Proc. IEEE Int. Conf. Computer Vision (ICCV), 2017.

[6] N. Dalal and B. Triggs, “HOGs of Oriented Gradients for Human Detection,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), 2005, pp. 886–893.

[7] R. Girshick, “Fast R-CNN,” in Proc. IEEE Int. Conf. Computer Vision (ICCV), 2015.

[8] M. Everingham et al., “The Pascal Visual Object Classes (VOC) Challenge,” Int. Journal of Computer Vision, vol. 88, no. 2, pp. 303–338, 2010.

[9] T.-Y. Lin et al., “Microsoft COCO: Common Objects in Context,” in Proc. European Conf. Computer Vision (ECCV), 2014.

[10] D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” arXiv preprint arXiv:1412.6980, 2014.

[11] I. Goodfellow, Y. Bengio, and A. Courville, “Deep Learning,” MIT Press, 2016.
Published
2026-04-22
How to Cite
Harsh Gupta, P Rajeev Siddarth, & Dr P. S. Thanigaivelu. (2026). Pothole Detection and Priority Ranking Using Deep Learning . IJRDO -Journal of Computer Science Engineering, 12(1), 11-20. https://doi.org/10.69980/cse.v12i1.6649
Issue
Vol. 12 No. 1 (2026): IJRDO - Journal of Computer Science Engineering (ISSN: 2456-1843)
Section
Articles

Copyright (c) 2026 IJRDO -Journal of Computer Science Engineering

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Author(s) and co-author(s) jointly and severally represent and warrant that the Article is original with the author(s) and does not infringe any copyright or violate any other right of any third parties, and that the Article has not been published elsewhere. Author(s) agree to the terms that the IJRDO Journal will have the full right to remove the published article on any misconduct found in the published article.