RDD2022: A multi‐national image dataset for automatic road damage detection

• Published in: Geoscience data journal 2024-10, Vol.11 (4), p.846-862
• Main Authors: Arya, Deeksha, Maeda, Hiroya, Ghosh, Sanjay Kumar, Toshniwal, Durga, Sekimoto, Yoshihide
• Format: Article
• Language: English
• Subjects: Aquatic reptiles; Artificial intelligence; Automation; Beavers; Big Data; Classification; Computer vision; Cracks; Damage; Damage detection; Data collection; Datasets; Deep learning; Earth science; Emergency preparedness; Flaw detection; global road damage detection; Guidebooks; intelligent transport system; Machine learning; Machinery condition monitoring; multi‐national data; Object recognition; R&D; Repair & maintenance; Research & development; Road conditions; Roads; Roads & highways; Sustainable development
• ISSN: 2049-6060; 2049-6060
• DOI: 10.1002/gdj3.260

The data article describes the Road Damage Dataset, RDD2022, encompassing of 47,420 road images from majorly six countries, Japan, India, the Czech Republic, Norway, the United States, and China. The dataset incorporates over 55,000 instances of road damage, specifically longitudinal cracks, transverse cracks, alligator cracks, and potholes. Designed to facilitate the development of deep learning methodologies for automated road damage detection and classification, RDD2022 was unveiled as part of the Crowd sensing‐based Road Damage Detection Challenge (CRDDC'2022), with a major contribution from the challenge winners. This challenge garnered global participation, urging researchers to propose solutions for automatic road damage detection in multiple countries. A noteworthy outcome of CRDDC'2022 was the emergence of a top‐performing model achieving a remarkable F1 Score of 76.9% for road damage detection in all six countries using RDD2022. This success underscores the dataset's practical applicability for municipalities and road agencies, enabling low‐cost, automatic monitoring of road conditions. Beyond its immediate utility, RDD2022 stands as a valuable benchmark for researchers in computer vision, geoscience, and machine learning, offering a rich resource for algorithmic evaluation in diverse image‐based applications, including classification and object detection. The latest big data cup, Optimized Road Damage Detection Challenge (ORDDC'2024), is also based on RDD2022, underscoring its continued relevance and pivotal role in current research and development endeavors. This paper introduces RDD2022, a versatile dataset collected from six countries for road damage detection. RDD2022 supports direct use for efficient, low‐cost road inspection and benchmarking deep learning algorithms. Augmentation potential lies in adding diverse images and damage categories, enhancing its utility. It holds immense promise for geoscience, computer vision, and machine learning applications, contributing to real‐world solutions.