<![CDATA[This scientific journal presents an approach to automatically detect road potholes from drone footage using a machine-learning model based on YOLOv5. The primary objective of this research is to develop a reliable and efficient system for road quality inspection. The proposed model achieves promising results with an F1 confidence of 0.83, Precision confidence of 0.96, Precision-Recall of 0.716, and Recall confidence of 0.8. The study aims to serve as a preliminary development toward the future implementation of road quality inspection. By leveraging drone footage and advanced machine learning techniques, the automated detection of potholes can significantly enhance the efficiency and accuracy of road maintenance efforts. Early detection and prompt repair of potholes can lead to improve road safety and reduce vehicle damage. Using drones and machine learning models allows for efficient monitoring and assessment of road infrastructure, contributing to sustainable transportation systems and minimizing the environmental impact of inefficient road maintenance. Moreover, this research contributes to the advancement of technology application in the field of environmental science. Overall, this study highlights the potential of YOLOv5-based machine learning models in automating the detection of road potholes from drone footage. The results demonstrate its effectiveness in accurately identifying and localizing potholes, paving the way for further advancements in road quality inspection and technology applications within the field of environmental science.]]>
