Real-Time Road Condition Detection and Mapping Using YOLOv11 and Built-In CarDashcam

Abstract. Road surface conditions decline due to heavy traffic, severe weather, and recurring utility works. Many road agencies still rely on manual windshield surveys and semi-automated inspections. These methods are time-consuming, difficult to scale, and labour-intensive. Advances in deep learning and the widespread availability of vehicle dashcams now offer new opportunities for low-cost, automated pavement assessments. This contribution presents a mobile, dashcam-based framework for detecting road-surface defects using the latest YOLOv11. This model is combined with geolocation tagging for spatial visualization. To test our YOLOv11 training model, we initially created a dataset at the University of the Fraser Valley campus. We manually annotated the dataset to identify crack fillings, crosswalk markings, speed bumps, lane markings, and other surface conditions. This was a prototype, with the aim of later training it to detect all road conditions, such as gravel, potholes, and uneven roads. To address variations in lighting and motion, augmentation techniques were applied. YOLOv11 achieved a mean average precision above 90% across all tested categories. This prototype demonstrates a practical, low-cost approach for real-time pavement monitoring. Future work includes expanding data collection, developing an operational dashboard for authorities, pinpointing exact GPS coordinates on maps with damaged road images, and evaluating model performance across different data sources, including models trained using Google Images. By producing actionable geospatial information, this system enables more efficient maintenance workflows. It provides a scalable pathway for municipalities seeking to modernize their road-condition assessments.