The GPR (Ground Penetrating Radar) conference in Hong Kong year 2016 marked the 30th anniversary of the initial meeting in Tifton, Georgia, USA on 1986. The conference has been being a bi-annual ...event and has been hosted by sixteen cities from four continents. Throughout these 30 years, researchers and practitioners witnessed the analog paper printout to digital era that enables very efficient collection, processing and 3D imaging of large amount of data required in GPR imaging in infrastructure. GPR has systematically progressed forward from “Locating and Testing” to “Imaging and Diagnosis” with the Holy Grail of ’Seeing the unseen’ becoming a reality. This paper reviews the latest development of the GPR’s primary infrastructure applications, namely buildings, pavements, bridges, tunnel liners, geotechnical and buried utilities. We review both the ability to assess structure as built character and the ability to indicate the state of deterioration. Finally, we outline the path to a more rigorous development in terms of standardization, accreditation, and procurement policy.
The polar ice sheet is vital for the global climate system, influencing the Antarctic ice sheet's mass balance, sea level rise, and Earth's surface energy. Accurate measurement of its thickness and ...internal structure is imperative for comprehending glacier evolution and climate change. Ground Penetrating Radar (GPR) is a high-resolution and non-invasive exploration instrument that is extensively used for glacier detection, but only limited geometric information can be identified in the GPR profile, making it difficult to capture the quantitative distribution of physical parameters. To image the fine dielectric parameters of GPR data acquired from the Amery Ice Shelf in East Antarctica in 2003, we propose a multi-trace GPR impedance inversion approach. This method utilizes the limited-memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS) algorithm for GPR data, coupled with weighted L2-norm total-variation multiplicative regularization (MR) to quantitatively estimate the dielectric parameters in the interior of the ice sheet. This scheme adaptively adapts regularization parameters, enhances vertical resolution, and suppresses noise. Additionally, considering lateral correlation, we integrate directional differences to enhance lateral continuity. After validating with complex test data, we apply the method to Amery Ice Shelf GPR data, quantitatively imaging the ice sheet's geometric and dielectric parameters. The inversion results reveal ice thickness, internal features, and the interface between freshwater ice and refrozen sea ice. This allowed us to determine the ablation and refreezing zone boundary at the bottom of ice shelf, providing insights into melting/freezing mechanisms, ice shelf stability, and simulating ice shelf interior dynamics.
•Introduction of a novel Ground Penetrating Radar (GPR) multi-trace impedance inversion scheme.•Successful inversion of GPR data for Amery Ice Sheet sounding data.•Imaging of relative permittivity through high-resolution inversion results.•Valuable insights into melting/freezing mechanisms and ice shelf stability provided by inversion results.
Non-destructive testing and characterization of internal vertical cracks are critical for road maintenance by ground penetrating radar (GPR). This paper describes a mask region-based convolutional ...neural network (R-CNN) that automatically detects and segments small cracks in asphalt pavement at the pixel level. Simulation using Gprmax software and field detection were performed to determine the crack features in GPR images of asphalt pavement and the relationship between the width of vertical cracks and their area in GPR images. Results showed that a 0.833 precision, 0.822 F1 score, 0.701 mean intersection-over-union (mIoU) and 4.2 frames per second (FPS) were achieved on 429 GPR images (1024×1024 pixels), and the mean error between the segmented crack width and the true values was 2.33%. The research results represent a further step toward accurately detecting and characterizing internal vertical cracks in asphalt pavement
•Detection frequency range for a 3D radar system was determined.•Vertical crack features in GPR images were determined.•Vertical cracks were detected using an improved Mask R-CNN model.
The radar equipment carried by the Chang'E-3 (CE-3) mission marked the first deployment of rover-mounted ground-penetrating radar (GPR) to observe the lunar surface. This provided an unparalleled ...opportunity for a high-resolution investigation into the fine structure of the lunar regolith. This paper has revealed the presence of multiple discrete layers within the top 4 meters of the lunar regolith using high-frequency radar data from the CE-3 Yutu rover. Subsequently, we have established realistic models of the lunar regolith to obtain the radar simulation calculated by Finite-difference time-domain (FDTD) technology. Thus, we compare the simulated radar data with actual observational data to comprehensively confirm the existence of multiple discrete layers within the lunar regolith. Taking into consideration the geological context of the CE-3 landing site and the principles of impact crater formation, we infer that the origin of the multiple discrete layers within the top 4 meters of the CE-3 landing site is likely the by-product of multiple depositions of ejecta from nearby small craters. Our findings suggest the possibility of the widespread existence of multiple discrete layers within the lunar regolith and emphasize the significant contribution of ejecta from small impact craters to the accumulation of local lunar regolith thickness on the Moon's surface.
The 2011 Tohoku‐oki tsunami caused large‐scale topographic changes along the Pacific coast of northeastern Japan. More than 10 years have passed since the tsunami waves struck the area. Today, ...because of reconstruction work, very few places exist where natural post‐tsunami topographic changes can be monitored continuously. For this study, the authors investigated topographic changes caused not only by the 2011 tsunami but also by natural and artificial activities during the 50 years before and after the tsunami based on aerial photographs, excavations and subsurface explorations using ground‐penetrating radar at the Osuka coast in Aomori prefecture, Japan. The site is rare because it is a protected area with few and superficial engineering activities, making it suitable for continuous observation of pre‐tsunami, syn‐tsunami and post‐tsunami topographic changes. The findings indicate that the 2011 tsunami waves generated large topographic changes: depositional and erosional features produced by the tsunami can be recognized, respectively, as tsunami deposits and erosional channels across the sand dunes. During the post‐tsunami phase, the sand volume at the coast quickly recovered naturally. Tsunami deposits and the erosional channels were well preserved underground even at 10 years after the event. However, dynamic movement of the dunes started after the tsunami. The shifting was attributable to the artificial clearing of coastal forests rather than the tsunami effects on the coast. Our results first indicate not only that the sedimentary features of paleo‐tsunamis but also the erosional features have some probability of being preserved in the subsurface of the beach and sand dunes at tsunami‐affected areas. Also, artificial activities such as deforestation are much more crucially undermining of the stability of the coastal geomorphology than the tsunami effects: the coast is now reaching a different status from its pre‐tsunami situation.
We conducted surveys of the area that suffered extensive deposition and erosion by the 2011 Tohoku‐oki tsunami. Tsunami deposits and the erosional channels were well preserved underground even at 10 years after the event. However, dynamic movement of the dunes started after the tsunami by the artificial clearing of coastal forests rather than the tsunami effects on the coast.
•Discussing advances of deep learning applications in GPR.•Discussing the existing issues of deep learning applications in GPR.•Comparing the architectures of deep leaning models exploiting GPR ...data.•Introducing the foundation of deep learning and GPR.
Deep learning has achieved state-of-the-art performance on signal and image processing. Due to the remarkable success, it has been applied in more challenging tasks, such as ground-penetrating radar (GPR) testing in civil engineering. This paper reviews methods involving deep leaning and GPR for civil engineering inspection and provides a classification based on the data types that they exploit. Based on the results of a comparison study, we conclude that methods using A-scan data slightly surpass the models using B- and C-scan data, though C-scan data is maybe the most promising in the further thanks to its complete space information. Two current limitations of deep learning exploiting GPR are its dependence on big data and overconfident decision-making. Therefore, benchmark GPR data sets and cautious deep learning are required.
Time-delay estimation (TDE) of pavement using ground penetrating radar (GPR) is an important task in the field of civil engineering. However, when dealing with GPR waves at the centimeter scale, ...there are challenges in real-time processing the coherent, overlapping backscattered echoes within limited bandwidth. To address these problems, this paper proposes a TDE method for GPR signals using co-prime sampling strategy via atomic norm minimization (ANM). The conventional uniform frequency sampling strategy requires lengthy data acquisition time and large data storage in GPR system. To address these problems, the co-prime sampling strategy is adopted in this paper to lower the sampling rate and lighten the hardware burden. The "holes" in co-prime sampling are filled by solving the ANM problem, where the reconstructed Hermitian Toeplitz matrix can be used for decorrelation without loss of degrees of freedom (DoFs). By selecting multiple-measurement vectors (MMV) directly from the second statistics of co-prime sampling signal, the scale of the semidefinite programing (SDP) of ANM is reduced. Moreover, we propose a 2-level ANM by recovering two Hermitian Toeplitz matrices instead of one Toeplitz matrix and one Hermitian matrix, reducing further the complexity of the proposed method. Numerical and experimental results show the superiority of the proposed method in terms of temporal resolution, estimation accuracy, and computational complexity.
Detecting buried objects from ground penetrating radar (GPR) profiles often requires manual interaction and plenty of time. This paper presents an automatic scheme for buried objects detection and ...localization. First, a trained deep learning framework — Faster R-CNN with data augmentation strategy is applied to identify hyperbolic signatures from a gray GPR B-scan image, which is capable of not only recognizing whether a B-scan profile contains traces of buried object, but also detecting candidate hyperbola region. Then, the detected rectangle region is extracted and transformed to a binary image, a novel double cluster seeking estimate (DCSE) algorithm is proposed to separate object point cluster from each other and enable the identification of hyperbolic signatures. Subsequently, a column-based transverse filter points (CTFP) method is utilized to extract hyperbola fitting points automatically from the validated point cluster. Downward opening hyperbola fitting is carried out and their respective peaks are obtained finally. The proposed scheme is able to extract information from GPR B-scan images automatically and efficiently; it is validated significant performance in the analysis of synthetic and on-site GPR data sets.
•Faster R-CNN with data augmentation strategy is applied to pick hyperbola region from the gray GPR B-scan image.•A novel double cluster seeking estimate (DCSE) algorithm is proposed to separate object point cluster from each other.•A column-based transverse filter points (CTFP) method is utilized to extract hyperbola fitting points automatically.•The experiments show significant performance in the analysis of synthetic and on-site GPR data sets.
To better reconstruct underground targets based on ground-penetrating radar (GPR) data, this paper proposes a joint physics and data driven full-waveform inversion (PDD-FWI) scheme. This scheme ...combines a physics-based non-iterative approach and a data-driven deep neural network (DNN) to reconstruct target location, shape and permittivity accurately. Firstly, the normalized range migration algorithm (RMA) is introduced to extract the target contour and location information, which not only improves the robustness of the proposed scheme, but also ensures adaptability to different GPR equipment. Then, the GPR dielectric target reconstruction network (GPRDtrNet) is developed based on the improved U-net structure, including reducing network layers and adding multi-scale additive spatial attention gates and skip-connection structures. Compared with previous DNN-based reconstruction methods, GPRDtrNet has the advantages of small data requirement, high accuracy, strong generalization and noise tolerance. Finally, the simulated and real dataset containing kinds of targets is constructed to train and test GPRDtrNet. The results show that the proposed method can reconstruct underground dielectric targets accurately with high robustness and noise tolerance.
