We are pleased to announce the open-source release of a comprehensive Ground Penetrating Radar (GPR) dataset collected by Professor Hai Liu and his team at Guangzhou University. This dataset encompasses a wide range of subsurface conditions and structures, including tunnel linings, underground pipelines, and reinforced concrete components. Our objective is to make these data publicly available in order to encourage academic and industrial researchers to explore and advance GPR methods for subsurface imaging and defect detection. 1.OverviewThis dataset contains raw GPR data collected under various real-world conditions. Specifically, it includes: Tunnel Lining Data: GPR scans collected from the Liangjiaying, Pingdingshan, and Niujianzi tunnels, covering scans of different sections within the tunnels. Folders labeled GD, YBQ, YG, YGJ, ZBQ, and ZGJ correspond to the tunnel’s crown, right wall, invert, right shoulder, left wall, and left shoulder, respectively. Underground Pipeline Data: Collected from the University Town in Guangzhou, showcasing measurements of various pipeline layouts and depths. Rebar Data: Collected from a residential area in Foshan, recording GPR signals from reinforced concrete structures with varying rebar densities and orientations. 2.Data Collection and Organization Instrumentation: Data were collected using commercial GPR systems operating at various frequencies to accommodate different penetration depths and resolutions. Format: The raw GPR data files are provided in .dt data format used by IDS GeoRadar equipment manufacturer. 3.Potential ApplicationsResearchers can utilize this dataset for a wide range of applications, including but not limited to: Subsurface Imaging: Optimizing imaging algorithms, such as reverse-time migration, for detecting cavities and defects behind tunnel linings. Antenna Parameter Evaluation: Studying optimal antenna configurations to improve the detection of hidden features and anomalies within tunnels. Concrete Inspection: Identifying and locating rebar, and evaluating the effects of different rebar configurations on the structure. Machine Learning and Deep Learning: Training and testing AI models for the automated detection, segmentation, and classification of subsurface features. 4. Related PublicationsWhen using this dataset in your research, please cite the following publications that inspired its collection and design: Detection and LocalizationLiu, H., Lin, C., Cui, J., Fan, L., Xie, X., & Spencer, B. F. (2020). Detection and localization of rebar in concrete by deep learning using ground penetrating radar. Automation in Construction, 118, 103279. Antenna EvaluationLiu, H., Yue, Y., Lai, S., Meng, X., Du, Y., Cui, J., & Spencer, B. F. (2023). Evaluation of the antenna parameters for inspection of hidden defects behind a reinforced shield tunnel using GPR. Tunnelling and Underground Space Technology, 140, 105265. Reverse-Time MigrationLiu, H., Yue, Y., Lian, Y., Meng, X., Du, Y., & Cui, J. (2024). Reverse-time migration of GPR data for imaging cavities behind a reinforced shield tunnel. Tunnelling and Underground Space Technology, 146, 105649. Full Waveform InversionWang, X., Chen, J., & Liu, H. (2024). Enhancing Full Waveform Inversion of Field GPR Data: A Source-Independent Approach with Dynamic Reference Selection via SE-Wave-U-Net. Geophysics, 90(1), 1-89. These references provide insights into advanced imaging techniques, antenna configurations, and machine learning approaches that can be further explored or extended using the dataset. 5.Data Access and License Availability: The dataset is hosted on an open-access repository. Please refer to the link provided on the project webpage for direct downloads and additional documentation. License: This dataset is released under a permissive open-source license to facilitate collaboration and knowledge sharing. Users are free to use, modify, and distribute the data, provided they comply with the terms of the license. 6.How to CiteWhen using or referencing this dataset, please cite both the dataset (according to the citation format provided in the repository) and the relevant publications listed above. Proper citation of our work and this dataset will help us continue to share resources and foster further advancements in GPR research. ContactFor any questions, feedback, or collaboration inquiries, please contact Professor Hai Liu's team at Guangzhou University. We look forward to seeing the innovative research and applications that emerge from this dataset.Official Team Website: http://www.gzhuybqxtc.com/col.jsp?id=101E-mail (Professor Hai Liu): hliu@gzhu.edu.cn AcknowledgmentsWe extend our gratitude to all the team members and collaborators who contributed to the data collection, annotation, and validation. We hope this resource will support your research efforts and inspire the future development of GPR technologies.