YOLOv11-Based AI-Driven Road Defect Detection, Severity Assessment, and Repair Cost Estimation Using OpenCV

This paper proposes an end-to-end system where a citizen will submit a picture of the road defect and it will get connected to the relevant Municipal Officer. YOLOv11 detects and classifies, OpenCV contour pipeline gives an approximation of the measurement, rule-based model assigns severity, parametric model estimates the repair cost, and the point-in-polygon method is used to determine the responsible ward and officer. These components are embedded in a web application, written in Flask, Node.js, React.js, and using MySQL, which citizens will then use without signing up for it. The key element is the integration of these stages and not the detector itself. However, with many of the systems in place, all the information is limited to the drawing of a bounding box, whereas the municipal application requires information on measurement, severity, cost, jurisdiction and assignment information as well. These are only placeholders as model training and deployment in the field are ongoing. The following steps involve recording the benchmarked performance outcomes, setting severity levels according to expert calibration and comparing costs with actual repair costs.

1. Khanam, R., Hussain, M.: YOLOv11: An overview of the key architectural enhancements. arXiv:2410.17725 (2024) 2. Ultralytics: YOLO11 Documentation and Configuration Reference. https://docs.ultralytics.com/usage/cfg (accessed 2026) 3. Maeda, H., Sekimoto, Y., Seto, T., Kashiyama, T., Omata, H.: Road damage detection and classification using deep neural networks with smartphone images. Computer-Aided Civil and Infrastructure Engineering 33(12), 1127–1141 (2018) 4. Arya, D., Maeda, H., Ghosh, S.K., Toshniwal, D., Sekimoto, Y.: RDD2022: A multi-national image dataset for automatic road damage detection. arXiv:2209.08538 (2022) 5. Arya, D., Maeda, H., Ghosh, S.K., Toshniwal, D., Omata, H., Kashiyama, T., Sekimoto, Y.: Crowdsensing-based road damage detection challenge (CRDDC’2022). In: 2022 IEEE International Conference on Big Data, pp. 6378–6386 (2022) 6. Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: Proc. IEEE CVPR, pp. 779–788 (2016) 7. Terven, J., Córdova-Esparza, D.M., Romero-González, J.A.: A comprehensive review of YOLO architectures. Machine Learning and Knowledge Extraction 5(4), 1680–1716 (2023) 8. Zhang, L., Deng, Y., Zou, Y.: Automatic road damage recognition based on improved YOLOv11 with multi-scale feature extraction and fusion attention mechanism. PLOS ONE (2025) 9. Zhou, H., Chen, S., Yan, G., Huang, Y., Lu, Z.: An intelligent road damage detection method based on YOLOv7-tiny. In: 2023 IEEE ICCT, pp. 286–290 (2023) 10. Sathvik, M., Saranya, G., Karpagaselvi, S.: An intelligent CNN-based potholes detection using YOLO-v7. In: 2022 ICACRS, pp. 813–819 (2022)