Research on Pavement Crack Detection Based on Deep Learning and UAV Aerial Images

Pavement distress detection forms a fundamental part of highway maintenance, among which cracks are the most common type of defect. Cracks not only compromise the structural integrity and aesthetic appearance of the pavement, but may also pose serious risks to driving safety. Traditional inspection methods often suffer from low efficiency, high labor costs, insufficient accuracy, and potential disruptions to traffic flow, highlighting the urgent need for more advanced alternatives. With the rapid advancement of computer vision and deep learning technologies, automated approaches based on object detection have been widely adopted in pavement distress inspection. Simultaneously, unmanned aerial vehicles (UAVs) have demonstrated significant potential in pavement image acquisition due to their high mobility, low cost, and ease of operation. This paper proposes a novel pavement crack detection method that integrates deep learning algorithms with UAV aerial imagery, aiming to achieve automated crack identification and precise localization. The proposed approach significantly improves detection efficiency and accuracy, offering strong engineering adaptability and promising potential for widespread application. It provides an effective technical foundation for the intelligent development of road maintenance