Search Results (92)

Tunnel lining defects pose significant threats to structural integrity and operational safety. Traditional image processing and machine learning methods often suffer from limited accuracy and poor generalization under complex backgrounds. To address these limitations, this study proposes a lightweight intelligent recognition method based on You Only Look Once version 11 nano (YOLOv11n) for Ground Penetrating Radar (GPR) images of tunnel linings. The backbone is replaced with Mobile Network Version 3 (MobileNetV3) to reduce parameters and Floating Point Operations (FLOPs), while depthwise separable convolution and a streamlined Compressed 2-Stage Fused-Lite (C2f-Lite) structure are integrated into the Neck to further decrease computational overhead. Channel mapping layers are employed to ensure smooth feature transfer, and selective use of Squeeze-and-Excitation (SE) attention and Hard-Swish (H-swish) activation balances detection accuracy with efficiency. Evaluated on a low-power mobile workstation acting as an edge-precursor proxy platform, experimental results demonstrate that the improved YOLOv11n_MobileNetV3 model achieves high accuracy with a mean Average Precision (mAP) at 0.5 of 94.4% and mAP@0.5:0.95 of 62.4%, low computational cost of 4.7 Giga Floating Point Operations (GFLOPs), and fast inference speed of 45 Frames Per Second (FPS). Comparative analysis further confirms its superior balance of detection performance and efficiency over YOLO version 5 (YOLOv5) and YOLO version 8 (YOLOv8) baselines. The proposed approach provides a highly optimized, edge-oriented engineering solution for real-time tunnel lining defect inspection, establishing strong structural and theoretical feasibility for future deployment in embedded systems. Full article

The reinforced concrete structures of many bridges and buildings in Taiwan are over 30 years old. Seismic retrofitting of these structures requires an accurate assessment of reinforcement configuration and corrosion conditions to ensure structural safety and seismic performance. In this study, a 1 GHz ground-penetrating radar (GPR) antenna was used to scan reflected signals from single- and double-row reinforcing bars embedded in concrete. Based on established principles reported in previous studies, detailed analyses were conducted, including the use of the approximate circumference method to estimate reinforcing bar dimensions and the determination of spacing between double-row reinforcing bars (6–8 cm). The synthetic aperture focusing technique was first applied to process the original GPR data matrix. Subsequently, physical parameters related to interface diffraction, such as the perimeter S of the reinforcing bar, were extracted using the dielectric constant of the material interface, the calculated power reflection coefficient, and the First Fresnel Zone. These approaches enabled more accurate estimation of reinforcing bar dimensions (e.g., equivalent to #3 bar size) and improved resolution of spacing between double-row reinforcing bars to 3–6 cm. The results demonstrate that using the synthetic aperture focusing technique to process GPR data enhances the ability to determine reinforcing bar dimensions, interpret bar spacing, and improve imaging resolution, thereby providing a reliable reference for the safety assessment of reinforced concrete structures. Full article

Investigating slope deformation in densely vegetated or remote areas is a major challenge for slope stability assessment. This study introduces and validates an integrated UAV-borne low-frequency Ground Penetrating Radar (UAV-GPR) and LiDAR methodology to characterize an unstable slope in Melizzano, Southern Italy. Radar data were acquired along an east–west transect at ~1 m above ground level, while high-resolution LiDAR were used to generate a detailed Digital Terrain Model for topographic correction and geomorphological analysis. The processed radargram images subsurface features down to ~15 m, revealing a laterally continuous high-amplitude reflector at ~10 m, interpreted as a key main sliding surface. Chaotic reflections above this interface indicate heterogeneous deposits associated with gravitational deformation, while more homogeneous reflections below correspond to stable geological units. The geometry of the reflector suggests a compound landslide mechanism. Borehole data validate the geophysical interpretation, showing depth discrepancies lower than 2 m. The integration of UAV-GPR and LiDAR enables a reliable correlation between surface morphology and subsurface structures. This non-invasive, spatially continuous approach provides an effective framework for subsurface characterization and for improving the interpretation of landslide geometry and internal structure in challenging environments. This study demonstrates the capability of low-frequency UAV-borne GPR to detect deep-seated sliding surfaces (>10 m) in vegetated environments when integrated with high-resolution LiDAR topography. Full article

Ballast compaction is a critical factor influencing ballast bed condition and the operational safety of heavy-haul railways. However, existing quantitative evaluation methods often suffer from overly idealized simulation models and limitations in signal processing and assessment frameworks. To address these issues, this study proposes a quantitative analysis approach for ballast compaction by integrating non-uniform medium simulation modeling, wavelet–Singular Value Decomposition (SVD) joint denoising, frequency–wavenumber (F-K) migration imaging, and wavelet packet decomposition (WPD)-based feature extraction. Forward simulations were conducted based on the constructed model, and the proposed methodology was validated using 1.5 GHz (gigahertz, 1 GHz = ${10}^9$ Hz) ground penetrating radar (GPR) data acquired from compaction experiments. The results demonstrate that wavelet–SVD joint denoising effectively suppresses deep coherent noise caused by strong reflections from sleepers, significantly enhancing the identification of deep effective signals and ensuring accurate localization and feature extraction of compaction zones. The Geometric Mean of WPD High/Low-Frequency Energy Ratio (GMHLFER) exhibits a strong positive correlation with the degree of compaction. In simulations, as the proportion of compacted material increased from $9\%$ to $21\%$, the GMHLFER rose from 21.555 to 26.581. In field tests, the value increased from 22.012 to 26.012 as compaction severity progressed from slight to severe, demonstrating stable responses across full-gradient compaction conditions and indicating high robustness and sensitivity. The proposed method provides an effective approach for quantitative characterization of ballast compaction in heavy-haul railways, and offers a promising technical pathway for intelligent inspection and condition assessment of railway ballast beds. Full article

Ground-penetrating radar (GPR) is a key non-destructive technique for subsurface reconstruction, widely valued for its ability to image buried structures without disruption. Among its various implementations, vehicle-mounted GPR has emerged as particularly suitable for highway tunnel assessment due to its rapid non-contact operation. However, current systems are often constrained by closely spaced antennas that generate strong direct coupling and consequently limit detection depth. To mitigate this issue, this paper proposes an antenna decoupling method based on coherent difference imaging. A differential decoupling model is first established to characterize the relationship between conventional transceiver signals and the derived differential signals, explicitly accounting for parameters such as antenna height and target depth. Furthermore, a coherent difference imaging algorithm is developed, employing a sliding-window coherence process to resolve dual-peak artifacts and restore focused target images. Simulations validate consistent performance across varying antenna heights, while experiments demonstrate over 37.2 dB isolation in the 1–3 GHz band and markedly improved imaging focus compared to conventional configurations, thereby enhancing buried target detection and supporting reliable data interpretation. Full article

The Bahamas is an ideal location for studying the calcium carbonate sedimentation of Holocene strandplains in relation to seaward progradation. We use ground-penetrating radar (GPR) to image and interpret the fine-scale stratigraphy of three carbonate strandplains on Crooked Island, The Bahamas. GPR has been extensively used to analyse the interiors of clastic strandplain deposits worldwide, while tropical carbonate settings have received less attention. Due to the lack of outcrops in our study area on Crooked Island, we validate the interpretation of the 2D profiles by comparing them with a 3D GPR data volume collected adjacent to and over a Pleistocene aeolianite outcrop on San Salvador Island, where porosity layering can be directly observed. Data processing employed state-of-the-art techniques adapted from the petroleum industry to enhance the visualisation of reflection amplitude on the GPR images. Our data support a model in which the progradation of carbonate sediment preserved in strandplains was deposited through a combination of storm processes and gradual sediment progradation. Full article

The detection of subsurface stromatolites remains challenging due to their complex morphology and heterogeneous composition. This study assesses the combined application of Electrical Resistivity Tomography (ERT) and Ground Penetrating Radar (GPR) for identifying microbialites in two contrasting geological and climatic settings: the Neoproterozoic Salitre Formation in Brazil and the Mesozoic microbialite-bearing limestones in northern Spain. High-resolution ERT profiles processed with raster-based blob detection algorithms revealed subcircular high-resistivity anomalies consistent with the studied microbialite morphologies, with strong resistivity contrasts observed between microbialites and host matrices despite variations in absolute values linked to lithology and soil moisture. In parallel, GPR surveys analyzed with a peak detection algorithm delineated domal reflectors and clusters of high-amplitude reflections that directly captured the internal architecture of stromatolitic buildups. With decimetric vertical resolution, GPR offered unrivaled insights into internal morphology, complementing the broader-scale imaging capacity of ERT. The complementary strengths of both methods are clear: ERT excels at mapping distribution and stratigraphic context, while GPR provides unparalleled resolution of internal structures. Crucially, this work advances previous efforts by explicitly demonstrating that integrated ERT-GPR approaches, when combined with algorithm-based interpretation, can resolve microbialite morphology, distribution and internal architecture with a level of objectivity not previously achieved. Beyond methodological refinement, these findings open new avenues for reconstructing microbialite development and preservation in ancient carbonate systems and hold strong potential for application in other geological contexts where complex carbonate structures challenge traditional geophysical imaging. Full article

This paper presents DeepMask-GPR, a novel deep learning framework for automatic detection and geometric estimation of buried cylindrical objects in ground-penetrating radar (GPR) B-scan images. Built upon Mask R-CNN, the proposed method integrates hyperbola detection, apex localization, and real-world coordinate mapping in an end-to-end architecture. A curvature-enhanced dual-channel input improves the visibility of weak hyperbolic patterns, while a quadratic regression loss guides the network to recover precise geometric parameters. DeepMask-GPR eliminates the need for raw signal data or manual post-processing, enabling robust and scalable deployment in field scenarios. On two public datasets, DeepMask-GPR achieves consistently higher TPR/IoU for spatial localization than baselines. On an in-house B-scan set, it attains low MAE/RMSE for radius estimation. Full article

Ground-penetrating radar (GPR) has been proven effective for detecting subsurface pipes in a nondestructive way, typically with manual processing and decision-making. However, existing automatic models for segmenting the target hyperbolas often lack generalization across different pipe radii, varying subsurface media, and complex field conditions. This is especially reflected in B-scans with diverse or small-scale hyperbolas, often accompanied by cluttered and irregular noise. In this paper, an automatic preprocessing model is proposed to enhance the interpretation of B-scans under challenging conditions. The model includes a ground reflection removal algorithm (GRRA), the data gravitational force enhancement (DGFE) method, and a global–local thresholding technique consisting of dilation-based local thresholding and segmentation (DLTS). First, a frequency-domain filter based on the fast Fourier transform and a spatial filter are applied to the raw B-scan to remove obstructive ground reflection strips. Owing to the minimal intensity differences among the target hyperbola, multiples, and background, the DGFE approach is introduced to amplify the main body of the hyperbola, distinguishing it from the noise. Finally, the target hyperbola is extracted from the grayscale image by an integrated thresholding approach. The approach initially employs global thresholding to eliminate all information except for part of the hyperbola, followed by DLTS, which uses a dilation operation with local thresholding to fully segment the hyperbola. The proposed model is evaluated on two distinct datasets and compared with several state-of-the-art methods. The results demonstrate its effectiveness, particularly in terms of cross-dataset generalization. Full article

The deployment of traditional ground-penetrating radar (GPR) systems for ice detection on steep terrain presents substantial safety challenges for ground crews due to inaccessibility and hazardous working conditions. However, airborne GPR (AGPR) and radio echo sounding (RES) provide solutions to these difficulties. Assuming that ice is homogeneous, we introduce a perspective back-projection algorithm designed to process AGPR or RES data that directly searches for unobstructed refracted electromagnetic (EM) wave paths and focuses EM energy below the surface by computing path-specific travel times. The results from the 2D and 3D imaging tests indicate that the perspective back-projection algorithm can accurately image the ice–rock interface. However, Snell’s Law suggests that part of the energy may fail to propagate through the air–ice interface and reach either the ice–rock interface or the receivers in scenarios where the incident angle of an EM wave exceeds a certain threshold. This energy deficit can hinder the perspective back-projection algorithm from accurately imaging such ice–rock interfaces. Despite these limitations, the perspective back-projection algorithm remains a promising tool for imaging sub-ice interfaces in AGPR and RES ice detection. Full article

Hyperbolic reflection features from ground-penetrating radar (GPR) data have been recognized as essential indicators for detecting hidden defects in the asphalt pavement. Computer vision and deep learning algorithms have been developed to detect and enhance the hyperbolic features of hidden defects. However, migrating existing hyperbolic feature detection methods using raw GPR data results in inaccurate predictions. Pre-processing raw GPR data using straightforward image processing methods could enhance the reflection features for fast and accurate hyperbolic detection during real-time GPR measurements. This study proposed accessible and straightforward image processing methods as GPR data preprocessing steps (such as the Sobel edge detector and histogram equalization) to assist existing computer vision algorithms for reflection feature enhancement during the GPR survey. Field tests were conducted, and several image processing methods with existing standard image processing libraries were applied. The proposed regions of the identified hyperbola signal-to-noise ratio (RIHSNR) were used to quantify the enhancement performance of hyperbolic feature detectability. Applying Sobel edge detection and Otsu’s thresholding to GPR data significantly improves detection accuracy and speed: mAP@0.5 rises from 0.65 to 0.85 for Faster R-CNN and from 0.72 to 0.88 for CBAM-YOLOv8 using the proposed computer vision methods as preprocessing steps. At the same time, inference time drops to 30 ms and 25 ms, respectively. Full article

Ground-penetrating radar (GPR) is an important geophysical technique in subsurface detection. However, traditional numerical simulation methods such as finite-difference time-domain (FDTD) face challenges in accurately simulating complex heterogeneous mediums in real-world scenarios due to the difficulty of obtaining precise medium distribution information and high computational costs. Meanwhile, deep learning methods require excessive prior information, which limits their application. To address these issues, this paper proposes a novel two-step forward modeling strategy for GPR data of metal pipes. The first step employs the proposed Polarization Self-Attention Image Translation network (PSA-ITnet) for image translation, which is inspired by the process where a neural network model “understands” image content and “rewrites” it according to specified rules. It converts scene layout images (cross-sectional schematics depicting geometric details such as the size and spatial distribution of underground buried metal pipes and their surrounding medium) into simulated clutter-free GPR B-scan images. By integrating the polarized self-attention (PSA) mechanism into the Unet generator, PSA-ITnet can capture long-range dependencies, enhancing its understanding of the longitudinal time-delay property in GPR B-scan images. which is crucial for accurately generating hyperbolic signatures of metal pipes in simulated data. The second step uses the Polarization Self-Attention Style Transfer network (PSA-STnet) for style transfer, which transforms the simulated clutter-free images into data matching the distribution and characteristics of a real-world underground heterogeneous medium under unsupervised conditions while retaining target information. This step bridges the gap between ideal simulations and actual GPR data. Simulation experiments confirm that PSA-ITnet outperforms traditional methods in image translation, and PSA-STnet shows superiority in style transfer. Real-world experiments in a complex bridge support structure scenario further verify the method’s practicability and robustness. Compared to FDTD, the proposed strategy is capable of generating GPR data matching real-world subsurface heterogeneous medium distributions from scene layout models, significantly reducing time costs and providing an efficient solution for GPR data simulation and analysis. Full article

Accurate detection of reinforcing bars (rebars) in concrete structures using ground penetrating radar (GPR) is crucial for effective structural evaluation but remains challenging, particularly when asphalt overlays compromise signal clarity. This study evaluates the performance of deep learning-based rebar detection using the You Only Look Once version 8 (YOLOv8) object detection model across three GPR datasets categorized as clear, interfering, and blurry. Models trained on each category were applied across varying conditions to assess generalization and robustness. A filtering algorithm was introduced to eliminate redundant and overlapping detections, thereby significantly improving the accuracy of YOLOv8-based predictions. The YOLOv8 approach outperforms traditional analytical techniques, especially under noisy or complex scenarios. In blurry GPR images where analytical methods fail, the filtered YOLOv8 model accurately detects rebar with a count that closely matches the ground truth. Across different datasets, the YOLOv8 approach demonstrates improved consistency in both location and quantity estimation, with filtered predictions correcting substantial over-detection seen in raw outputs. The study presents a practical framework for applying deep learning to GPR data, enhancing the reliability of rebar detection under diverse field testing and evaluation conditions. The findings highlight the importance of developing tailored training datasets and post-processing strategies when deploying AI tools for in-service bridge inspections. Full article

Localizing ground-penetrating radar (LGPR) benefits from deep subsurface coupling, ensuring robustness against surface variations and adverse weather. While LGPR is widely recognized as the complement of existing vehicle localization methods, its reliance on prior maps introduces significant challenges. Channel misalignment during traversal positioning and time-dimension distortion caused by non-uniform platform motion degrade matching accuracy. Furthermore, rain and snow conditions induce subsurface water-content variations that distort ground-penetrating radar (GPR) echoes, further complicating the localization process. To address these issues, we propose a weather-resilient adaptive spatio-temporal mask alignment algorithm for LGPR. The method employs adaptive alignment and dynamic time warping (DTW) strategies to sequentially resolve channel and time-dimension misalignments in GPR sequences, followed by calibration of GPR query sequences. Moreover, a multi-level discrete wavelet transform (MDWT) module enhances low-frequency GPR features while adaptive alignment along the channel dimension refines the signals and significantly improves localization accuracy under rain or snow. Additionally, a local matching DTW algorithm is introduced to perform robust temporal image-sequence alignment. Extensive experiments were conducted on both public LGPR datasets: GROUNDED and self-collected data covering five challenging scenarios. The results demonstrate superior localization accuracy and robustness compared to existing methods. Full article

Civil engineering structures with damage, defects, or subsurface utilities create a high-contrast exploration environment. These anomalies of interest exhibit different electromagnetic properties from the surrounding medium, and ground-penetrating radar (GPR) has the potential to accurately locate and map their three-dimensional (3D) distributions. However, full-waveform inversion (FWI) for GPR data struggles to simultaneously reconstruct high-resolution 3D images of both permittivity and conductivity models. Considering the magnitude and sensitivity disparities of the model parameters in the inversion of GPR data, this study proposes a 3D dual-parameter FWI algorithm for GPR with a composite model constraint strategy. It balances the gradient updates of permittivity and conductivity models through performing total variation (TV) regularization and minimum support gradient (MSG) regularization on different parameters in the inversion process. Numerical experiments show that TV regularization can optimize permittivity reconstruction, while MSG regularization is more suitable for conductivity inversion. The TV+MSG composite model constraint strategy improves the accuracy and stability of dual-parameter inversion, providing a robust solution for the 3D imaging of subsurface anomalies with high-contrast features. These outcomes offer researchers theoretical insights and a valuable reference when investigating scenarios with high-contrast environments. Full article