Article
Remote Sensing
Tess Luo, Song Zhu, Yiliminuer Yikeremu, Jiasong Zhu, John van Genderen
Summary: This study proposes a method for preliminary diagnosing large-scale culverts based on GPR wave attributes, which includes evaluating soil compactness, investigating soil moisture, and overlaying the results to pinpoint potential degradations. The method was validated through case studies on two long-distance culverts. The results indicate that this method can provide quick reference to improve GPR survey efficiency and reduce workload, facilitating large-scale GPR culvert surveys and safeguarding the water system.
GEO-SPATIAL INFORMATION SCIENCE
(2023)
Article
Environmental Sciences
Hui Wang, Shan Ouyang, Qinghua Liu, Kefei Liao, Lijun Zhou
Summary: Accurately estimating the relative permittivity of buried targets is crucial for reconstructing geological structures. This study proposes a method based on a deep neural network to effectively identify the dielectric properties of buried targets and improve the accuracy of estimating the relative permittivity.
Article
Robotics
Chieh Chou, Haifeng Li, Dezhen Song
Summary: This article discusses system and algorithmic developments for a sensing suite designed for surface and subsurface infrastructure inspection. It addresses a novel GPR-camera calibration problem and a synchronization-challenged sensor fusion problem, resulting in successful 3-D reconstruction and reduction of end-to-end distance error.
IEEE TRANSACTIONS ON ROBOTICS
(2021)
Article
Environmental Sciences
Gabbo P. H. Ching, Ray K. W. Chang, Tess X. H. Luo, Wallace W. L. Lai
Summary: A guidance system was developed in this study to help GPR operators conduct three-dimensional imaging surveys more efficiently, reducing survey time and improving image resolution.
Article
Computer Science, Interdisciplinary Applications
J. Hunziker, E. C. Slob, J. Irving
Summary: To address the computational cost issue in modeling GPR reflection data on glaciers, the authors propose a semi-analytical method based on the assumption of a homogeneous background medium. By representing scattering surfaces with planar elements and considering the antenna radiation pattern, this method can produce realistic 3D GPR data in a fast and memory efficient way. The algorithm is validated with an analytical solution for a layered model and applied to simulate radar data for the Otemma glacier in Switzerland.
COMPUTERS & GEOSCIENCES
(2023)
Article
Construction & Building Technology
Jiasong Zhu, Dingyi Zhao, Xianghuan Luo
Summary: Ground-penetrating radar (GPR) is an efficient method for diagnosing urban road defects, but its interpretation is complex. This study proposes an optimized YOLO-based framework for timely identification of road defects using GPR. Transfer learning and data augmentation are used to optimize the framework, and YOLOv5_s is found to perform the best. The framework is validated using real GPR data and provides comparable accuracy within seconds, benchmarking YOLOv5_s for timely road inspection.
ROAD MATERIALS AND PAVEMENT DESIGN
(2023)
Article
Geochemistry & Geophysics
Hai Liu, Zhijie Chen, Hantao Lu, Feng Han, Chao Liu, Jing Li, Jie Cui
Summary: In this letter, a modified migration algorithm is proposed for reconstructing the geometric structure of a subsurface object from ground penetrating radar (GPR) data. The algorithm corrects for the influence of the antenna radiation pattern in subsurface soil and has been verified through numerical, laboratory, and field experiments. The results demonstrate the superiority of the modified migration algorithm in suppressing undesired diffractive artifacts while preserving reflection amplitudes in the migrated images.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Construction & Building Technology
Hai Liu, Jingyang Zhong, Feng Ding, Xu Meng, Chao Liu, Jie Cui
Summary: A hybrid-polarization GPR system is proposed in this paper for detection and evaluation of rebar corrosion in concrete, providing more accurate recording and estimation of the corrosion status of rebars. Results from an accelerated corrosion experiment show that radar signals in the polarization channel perpendicular to the rebar orientation are more sensitive to corrosion compared to the commonly-used parallel polarization channel.
CONSTRUCTION AND BUILDING MATERIALS
(2022)
Article
Computer Science, Information Systems
Xin Zhang, Liangxiu Han, Mark Robinson, Anthony Gallagher
Summary: This paper introduces a deep learning framework based on Generative Adversarial Nets (GANs) to address the scarcity of GPR data by generating new training data, automatically learning features, and detecting underground objects with high accuracy. This approach outperforms existing methods and demonstrates good generalizability through cross-validation on independent datasets.
Article
Geosciences, Multidisciplinary
Yue Yu, Chi-Chih Chen
Summary: This paper presents an automatic buried unexploded ordnance (UXO) classification methodology based on H-alpha entropy-based polarimetric features extracted from wideband fully polarimetric ground penetrating radar data. The method establishes a background H-alpha feature to determine the classification boundary of random and isotropic scatters, and detects UXO-like elongated targets through the presence of sufficient H-alpha feature in the linear scattering classification zone. The performance of the proposed UXO classifier is evaluated using actual field data collected at the US government Fort Ord UXO test site.
JOURNAL OF APPLIED GEOPHYSICS
(2023)
Article
Computer Science, Information Systems
Maxwell M. Omwenga, Dalei Wu, Yu Liang, Li Yang, Dryver Huston, Tian Xia
Summary: This article introduces an autonomous cognitive GPR system enabled by deep reinforcement learning, which outperforms other systems in terms of detection accuracy and operating time in underground exploration.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Review
Forestry
Brunela Pollastrelli Rodrigues, Christopher Adam Senalik, Xi Wu, James Wacker
Summary: This paper is a review of studies on the use of ground penetrating radar (GPR) for wood structures, highlighting its advantages as an inspection tool and the gaps in knowledge for its practical application. Despite laboratory studies, the use of GPR on large wood structures remains limited. Key knowledge gaps include distinguishing internal feature types and identifying internal decay.
Article
Geochemistry & Geophysics
Bin Liu, Yuxiao Ren, Hanchi Liu, Hui Xu, Zhengfang Wang, Anthony G. Cohn, Peng Jiang
Summary: GPRInvNet is a deep neural network architecture for inverting subsurface structures from ground-penetrating radar (GPR) B-Scan data, capable of effectively reconstructing complex defects in tunnel linings. By fusing features from adjacent traces to enhance each trace and further condensing features, it achieves accurate spatial alignment.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Review
Construction & Building Technology
Tim Klewe, Christoph Strangfeld, Sabine Kruschwitz
Summary: This article provides an overview of various approaches to moisture measurements with Ground Penetrating Radar (GPR) in civil engineering, emphasizing signal features and discussing their limitations. The study encourages the consideration of approaches that combine different signal features for further developments in the field.
CONSTRUCTION AND BUILDING MATERIALS
(2021)
Article
Geochemistry & Geophysics
Yue Yu, Chi-Chih Chen
Summary: This method combines SP data collected from at least three arbitrary antenna orientations to produce FP scattering matrix for subsurface target classification. Analysis of errors in the proposed SP-derived FP scattering matrix components due to the presence of cross-polarization component in practical SP antennas and random noise is also provided. Simulation and measurement examples demonstrate the application of the proposed SP-to-FP method in GPR target classification.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)