Paper Title
Deep Learning-Driven Road Damage Detection Via UAV Imaging And Enhanced YOLO Frameworks

Abstract
The article offers a new application of drones and cutting-edge deep learning technologies to locate road damage automatically. Maintainments of road equipment is significant to ensure safe road use yet data gathering is risky and consumes a lot of labor doing it manually. To respond, we find the damage on the road many times more efficiently and precisely with the help of drones and AI. YOLOv5 and YOLOv7 are two of the newest algorithms that we employ in order to identify objects in UAV images. Its numerous trainings and tests on Chinese and Spanish datasets prove that YOLOv7 provides the highest number of correct results. YOLOv8 is also introduced to our research. It performs better than other algorithms when trained on road damage data and also demonstrates even more correct prediction. These findings demonstrate that deep learning and UAVs would be applicable in detecting road damage. This may bring more work in this field. Keywords - UAV, Road Damage Detection, Deep Learning, Object-Detection, YOLOV5, YOLOV7, and YOLOV8