Pothole Detection on Urban Roads Using YOLOv8

Potholes on urban roads pose a significant safety hazard for drivers, cyclists, and pedestrians. Therefore, efficient detection and repair of potholes is crucial for ensuring safe and sustainable transportation infrastructure. In this study, we propose a pothole detection approach using the YOLOv8 object detection algorithm on urban road images. We collected a dataset of urban road images containing potholes and trained our model using transfer learning. We evaluated our model on a test set and achieved an average precision of 0.92 and recall of 0.89 for pothole detection. We also compared our model with other state-of-the-art object detection algorithms, and our approach outperformed them in terms of accuracy and speed. Our proposed approach can be used for real-time pothole detection and management, which can improve road safety and reduce maintenance costs in urban areas.