URDD: An open dataset for urban roadway disease detection and classification

Urban traffic accidents have become more common in recent years due to the rising number of motorized vehicles, climate change, and outdated subsurface drainage systems. Traditional road disease detection methods involve collecting road data using ground-penetrating radar and manually analyzing the data. This process is time-consuming and subjective. Deep learning, especially convolutional neural networks (CNNs), has proven highly effective in image recognition and object detection. By applying these techniques to road disease detection, both the efficiency and accuracy of detection can be significantly improved. To support this, we have created a specialized road disease dataset designed for object detection and classification tasks. The release of this dataset aims to promote the use of artificial intelligence (AI) in autonomous road disease detection and classification, enhancing detection efficiency and contributing to better urban road maintenance and management.