Smart Vehicle-Mounted Pothole Detection and Mapping System with Real-Time Database Integration

The Smart Vehicle-Mounted Pothole Detection and Mapping System enhances road safety and maintenance by utilizing the YOLOv4- Tiny algorithm for real-time pothole detection via dash cam footage. Detected potholes are logged with GPS coordinates and transmitted to a Firebase database, which continuously updates a map of road conditions. A mobile app provides drivers with alerts when approaching a pothole within a user-defined radius, aiding in safe driving and helping authorities prioritize road repairs. The system was tested under various road conditions, achieving a mean Average Precision (mAP) of 80.7% for pothole detection. It was evaluated using different pothole sizes and environments, and the YOLOv4-Tiny model showed reliable detection with minimal false positives. The real-time notification feature effectively alerted drivers, while the cloud-based database provided accurate, actionable data for road maintenance. The system demonstrated efficient performance and scalability, making it a practical solution for improving both road safety and infrastructure upkeep.