GPR Imaging Research Articles

GPR Forward Modelling and Imaging Analysis for Boulder Detection

GPR Forward Modelling and Imaging Analysis for Boulder Detection

Ellipse Inversion Model for Estimating the Orientation and Radius of Pipes From GPR Image

Ellipse Inversion Model for Estimating the Orientation and Radius of Pipes From GPR Image

Detecting Subsurface Voids From GPR Images by 3-D Convolutional Neural Network Using 2-D Finite Difference Time Domain Method

In this article, an algorithm for detecting subsurface voids under the road from ground penetrating radar images is proposed. A multichannel radar system mounted on vehicle enables dense and highspeed monitoring. The novelty of the algorithm is a unique ElectroMagnetic simulation method and state-of-the-art deep learning technique to consider three-dimensional (3-D) reflection patterns of voids. To train deep learning models, 3-D reflection patterns were reproduced by 2-D finite difference time domain method to drastically reduce the calculation cost. Hyperboloid reflection patterns of voids were extracted by 3-D convolutional neural network (3D-CNN). The classification accuracy of 3D-CNN was up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$90$</tex-math></inline-formula> %, about 10% improvement compared to previous 2D-CNN to demonstrate the effectiveness of 3-D subsurface sensing and detection. The results were validated by real void measurement data. After applying trained 3D-CNN to radar data, regions of voids were plotted in a 3-D map, offering clear visualization of areas of voids.

A New Methodology for the Detection and Extraction of Hyperbolas in GPR Images

Reliable detection of underground infrastructure is essential for infrastructure modernization works, the implementation of BIM technology, and 3D cadasters. This requires shortening the time of data interpretation and the automation of the stage of selecting the objects. The main factor that influences the quality of radargrams is noise. The paper presents the method of data filtration with use of wavelet analyses and Gabor filtration. The authors were inspired to conduct the research by the fact that the interpretation and analysis of radargrams is time-consuming and by the wish to improve the accuracy of selection of the true objects by inexperienced operators. The authors proposed automated methods for the detection and classification of hyperboles in GPR images, which include the data filtration, detection, and classification of objects. The proposed object classification methodology based on the analytic hierarchy process method introduces a classification coefficient that takes into account the weights of the proposed conditions and weights of the coefficients. The effectiveness and quality of detection and classification of objects in radargrams were assessed. The proposed methods make it possible to shorten the time of the detection of objects. The developed hyperbola classification coefficients show promising results of the detection and classification of objects.

Ground-Penetrating Radar Imaging of Near-Surface Deformation along the Songino Active Fault in the Vicinity of Ulaanbaatar, Mongolia

The seismic activity observed in the vicinity of Ulaanbaatar (UB) capital city has been increased since 2005. Several active faults have been identified in the UB area. Most of the Mongolian population is concentrated around UB (1.5 million), which is the main political and economic center of the country. Hence, the study of seismic hazard is of first importance for the country. In this paper, we present the GPR results obtained on the Songino fault which is situated at 20 km west-southwest of UB at the northeast tip of Khustai fault. The combination of the morphotectonic, GPR and paleoseismological investigations brings essential information for seismic hazards assessments. The 2D GPR profiles are measured by using 250 and 500 MHz antennae and the topography using a differential GPS. An appropriate processing of the GPR data, including the topographic migration, allows us to bring out indirect characteristics of these faults. The objective is to identify near-surface geometry and coseismic deformation along the mapped fault. The 250 MHz GPR images of the Songino fault show the evolution of the sub-surface deformation mode induced by the arched geometry of the Songino fault. We observe a clear compressive structure at its NW section, strike slip at its central section and extensive structure in its SE part.

Multi-Trace Coherence Mapping and Data Fusion for Improved GPR Imaging

– Usually only GPR reflection amplitudes are imaged, while GPR data contain far more information. – Multi-trace GPR coherency visualisations offer new insights into structures otherwise hardly visible. – Image fusing permits the combined display of different GPR data attributes (amplitude and coherency).

The Fortress Beneath: Ground Penetrating Radar Imaging of the Citadel at Alcatraz: 1. A Guide for Interpretation

Ground-penetrating radar has emerged as a prominent non-destructive evaluation tool for the study of inaccessible subsurface elements of cultural heritage structures. Often of central interest is the desire to image the remains of a pre-existing historic structure that is located directly beneath a more recently built one. The interpretation of GPR images in such cases is usually difficult due to ambiguities caused by the presence of pervasive clutter, environmental noise, and overlapping target signatures. Sites with abundant ground truth allow for more confident interpretations and serve as a useful testbed to assist similar studies at other places, where little or no ground truth is available. This study reports GPR interpretations of structures belonging to the 19th century Citadel beneath the main prison cellhouse at Alcatraz. At this site, lidar scans, direct observations, and historical documents are available to facilitate identification of radar target signatures. A general interpretation of the acquired radargrams is made in this paper, while the companion paper presents more advanced analysis of target signatures based on curvelet image processing. This study points to the development of a radar facies classification scheme that is specific to cultural heritage investigations.

GPR Virtual Guidance System for Subsurface 3D Imaging

Three-dimensional GPR imaging requires evenly and densely distributed measurements, ideally collected without the need for ground surface markings, which is difficult to achieve in large-scale surveys. In this study, a guidance system was developed to guide the GPR operator to walk along a predesigned traverse, analogous to the flight path design of an airborne drone. The guidance system integrates an auto-track total station unit (ATTS), and by estimating the real-time offset angle and distance, guidance corrections can be provided to the operator in real time. There are two advantages: (1) reduced survey time as grid marking on the ground is no longer needed and (2) accurate positioning of each traverse. Lab and field experiments were conducted in order to validate the guidance system. The results show that with the guidance system, the survey paths were better defined and followed in terms of feature connectivity and resolution of images, and the C-scans generated were closer to the real subsurface world.

Automated detection of cracks in roads using ground penetrating radar

A critical assessment criterion for road condition assessment is the amount and severity of cracking. Cracking causes a loss of structural capacity (i.e. strength and stiffness) and if severe can cause water infiltration into the subbase and subgrade, potentially leading to more serious structural failure. The onset of cracking therefore accelerates the processes of deterioration of the road. To date the accurate and reliable automated detection of road surface cracking is still problematic.Ground Penetrating Radar (GPR) has the potential to detect cracks in materials as they cause changes in electromagnetic reflection. This paper proposes a method of automatic detection of road surface cracks using GPR and establishes the limits of its potential regarding the detectable crack size.Tests were conducted on asphalt pavement samples in a controlled laboratory environment. Significant changes in diffracted signal levels were observed in the GPR image when the image was analysed using a Singular Value Decomposition (SVD) algorithm for clutter reduction and a median filter to reduce noise. Moreover, a manifestly enhanced performance was achieved using a novel post-processing technique, which enabled the detection of cracks larger than 1.3 mm in width. The detected cracks were detectable irrespectively of the simulated physical conditions suggesting that the developed methodology is robust for use in practice. The experimental results indicate that the GPR system can be applied to automatically detect road cracking reliably in practice.

A Comparison of Linear Inverse Scattering Models for Contactless GPR Imaging

Ground-penetrating radars operating at a stand-off distance from the probed domain are attractive diagnostic tools in several contexts. Despite notable advantages offered in terms of noninvasiveness and flexibility, the usage of stand-off configurations makes the imaging process more difficult. Indeed, an accurate reconstruction of the subsurface/hidden scene requires accounting for the presence of the air–soil interface into the scattering model. This article deals with a comparison of three different linear models for describing the scattering phenomenon under a stand-off configuration, where the presence of the layered medium is explicitly accounted for in the scattering model. The first model is rigorous and based on the spectral domain representation of the incident field and Green’s function for the half-space scenario. The second model exploits a ray representation of the wave propagation from the radar to the target and vice versa by accounting for the refraction phenomenon at the interface. The third model is an approximated one, where the layered nature of the scenario is taken into account by means of an equivalent wavenumber, which allows a straightforward evaluation of the incident field and Green’s function. The resolution performance of each inversion approach is assessed by means of a numerical analysis. Moreover, data referred to an experiment carried out in controlled conditions are processed and reconstructions are provided to support the comparative analysis among the three imaging approaches.

Cloud‐based difference algorithm using big GPR data for roadbed damage detection

SummaryGround‐penetrating radar (GPR) technology is being widely used for urban roadbed damage detection. In this paper, we present the development of a cloud‐based difference detection method using big GPR data for roadbed damage detection. Unlike many processing methods relying on standalone servers, the proposed method can achieve high accuracy and processing efficiency by distributing the computation over a server cluster. The method includes a difference detection algorithm for GPR image enhancement and damage extraction, the Hadoop Distributed File System, and MapReduce distributed parallel computing framework for big GPR data parallel processing and interpretation. Moreover, to automatize roadbed damage classification, a criterion with seven grades is established based on the wave group shape, amplitude, phase, and attenuation characteristics. We validate the proposed method through a detection experiment that was conducted on the Fourth Ring Road in Beijing, China. The detection revealed 187 damage from the efficient processing of big GPR data, thus suggesting the ability of the proposed method to provide timely support for road safety in cities.

Shape classification of ground penetrating radar using discrete wavelet transform and principle component analysis

Ground penetrating radar is one of the promising non-destructive investigation for shallow subsurface exploration in locating buried utilities. However, interpreting hyperbolic signature of buried objects in GPR images remains a challenging task since the GPR signals are easily corrupted by environmental noise and cause misinterpretation of the size and geometry of subsurface object from the GPR raw profile. Therefore, this paper proposes Discrete Wavelet Transform (DWT) and principal component analysis (PCA) to classify geometry of buried object using k-nearest neighbour. (k-NN). The GPR images firstly being pre-processed. Then, the GPR images are decomposed using DWT into four sub-bands which are LL (Low-Low), LH (Low-High), HL (High-Low) and HH (High-High). The sub-bands LL or a coarse approximation coefficients was extracted as DWT features in order to classify the shape of buried objects. Since DWT features do contain high dimensional data, thus PCA is used to reduce the dimensional features from higher to lower space by linear transformation. The new projected features were then classify using k-NN classifier into four shapes which is cubic, cylinder, disc and sphere. A series of experiments have been conducted on extracted DWT and PCA features from hyperbolic signature of buried objects having different shapes. Based on the results, the proposed method had achieved the average recognition rate of 99.41%.

A dielectric horn antenna and lightweight radar system for material inspection

In this paper, we set up a lightweight, frequency-domain radar system using a handheld vector network analyser combined with air-coupled and dielectric-coupled horn antennas. A new positioning system with a binary barcode ruler, reading webcam and an algorithm for its recognition was developed. The radar system was tested on several media using both the air-coupled and dielectric-coupled antennas, namely, on a laboratory model containing sand and on a concrete block with a metallic reinforcement. The resulting GPR images demonstrated the good functioning of the radar system. In our experiments, both antennas and the whole lightweight radar system provided radar images which permitted to clearly detect the buried objects. Furthermore, the dielectric-coupled horn antenna showed better coupling with the investigated materials, resulting in higher-quality radar images.

Measurement of iron mass in concrete wall using ultra‐wide band radar

Here, it is aimed to measure and detect the iron mass of a concrete solid wall, column, floor by using of ultra-wide band radar. Electromagnetic waves boundary conditions, reflection, and refraction of waves that radiate in solid material are analysed. GPR imaging methodology and scanning types such as A, B, and C are analysed. Ultra-wide band radar imaging is used for detecting measuring the diameter of the iron material.

Clogging in constructed wetlands: Indirect estimation of medium porosity by analysis of ground-penetrating radar images

Identification of the degree of porous medium clogging in constructed wetlands (CWs) is a complex procedure because bed obstruction is a phenomenon that occurs in the subsurface and involves several factors. Ideal methods must allow for subsurface analysis and be non-invasive, what is the case of ground penetrating radar (GPR or georadar) used for soil characterization. Thus, this work aimed to evaluate the use of GPR for clogging characterization of two full-scale horizontal subsurface flow constructed wetlands (HSSF-CWs), one unit planted with cattail (Typha latifolia) and the other unit with no vegetation. Both units received municipal sewage (50 p.e. each) previously treated in an anaerobic reactor, had been in operation for seven years (at the time the GPR was used) and showed signs of heavy clogging, leading to surface flow. In order to produce a standard for GPR image (created by the response of the reflected wave in the passage through different media) identification and association of colors with the environmental conditions inside the medium (cleaner or more obstructed condition), a clean granular rock filter (in operation for only one month), similar to the unplanted unit, was used. Equations developed for indirectly estimating the porosity, based on the RGB (red, green and blue) color scale, indicated that the methodology was in agreement with the visual conditions of surface flow occurrence in the HSSF-CW, showing to be a suitable non-invasive method to characterize the advancement of clogging in CWs.

GPR imaging criteria

GPR has been widely acknowledged as an effective and efficient technique for imaging the subsurface. But the process of constructing 3D GPR images (C-scans) is still subjective and mainly relies upon the operator's knowledge and experience. This study reviews the parameters that affect GPR imaging quality: namely, profile spacing (PS), slice thickness (ST) and interpolations. Feature characteristics that have a crucial influence on imaging quality were also identified. Through conducting 25 carefully designed empirical experiments on concrete as well as subsurface structures, the relationship between 3D imaging parameters and feature characteristics were observed. A general workflow was derived for GPR C-scan generation, which is analogous to the typical signal processing steps used in 2D radargram signal processing (Jol, 2009). Empirical values in workflow were based on the retrieval of known ground-truth data and comparison with the processed images, i.e. the closest to reality. Unlike 2D processing, the workflow for 3D is feature-oriented and case-specific, and the proposed workflow gives guidelines on suitable ranges for 3 major parameters when used in a variety of applications. It was identified that feature shapes and the ratios of feature size to radar footprint are of vital importance. With the proposed flowchart, the often vague “survey experience” is parametrized and standardized, and the upper and lower limits governing the generation of objective and trustworthy 3D GPR images are defined. This workflow for GPR 3D slice imaging also paves the way for GPR feature extraction and change detection commonly adopted in remote sensing.

Geophysical Studies at the Horogoru Fortified Settlement in South Korea

The results of prospection studies at the medieval site Horogoru, in the Gyeonggi-do Province of South Korea, are described. Using ground-penetrating radar, the defense wall, built of tamped earth and masonry, was reconstructed. The analyses of GPR images and 3D-models of the wall were confi rmed and supplemented by archaeological excavations. Prospection studies in the central part of the site have enabled us to assess tentatively the thickness of the habitation layer and its preservation. Structures associated with various archaeological cultures were analyzed. The results of excavations demonstrated a relative reliability of GPR, which had revealed anomalies at various depths. However, an accurate and complete assessment of the outlines of most structures proved impossible, owing to repeated medieval rebuilding, peculiar accumulation processes, and modern disruption. The GPR analysis of the anomalies indicated several stages of habitation. Early features, dating to the Koguryo period (400–700 AD), include a reservoir and a well, and next to these, heaps of roof tiles. Late features, dating to the Koryo stage (1000–1200 AD), include seven buildings, a stone pavement, and pits with roof tiles. Overall, the results demonstrate the effi ciency of geophysical methods for the assessment of the site’s structure and of the preservation of its cultural layers.

Detection of buried objects in ground penetrating radar data using incremental nonnegative matrix factorization

ABSTRACTGround penetrating radar (GPR) is a popular tool for subsurface sensing and it is widely used for buried object detection. In this study, a new buried object detection method based on the modelling of A-scans by incremental nonnegative matrix factorization (INMF) is presented. The existing clutter in the GPR image is learned via nonnegative matrix factorization (NMF) and the resulting basis and encoding matrices are used in the initialization of the INMF method. Since clutter is learned by NMF in the initialization, the target is considered as an anomaly and a new A-scan containing target signal results in an increase in the error signal level permitting the detection of the target at this antenna location. The proposed method is applied to an original and noisy simulated dataset constructed by the electromagnetic simulation software gprMax as well as to a real dataset. The quantitative results based on receiver operating characteristic (ROC) curves and area under curves (AUC) obtained for the simulated dataset with different SNR levels show that there is an improvement around 5–7% in the detection rate.

Performance Analysis of Detecting buried pipelines in GPR images using Faster R-CNN

Performance Analysis of Detecting buried pipelines in GPR images using Faster R-CNN

Processing considerations and improved interpretation for ground‐penetrating radar imaging of a relict archaeological excavation unit

ABSTRACTA ground‐penetrating radar survey was carried out at The Great Temple Mound (Mound A), Ocmulgee National Monument, Georgia, USA to determine the location of buried archaeological features. Mound A is the largest pre‐historic mound at the site and its function remains unclear. High quality data and sufficient spatial sampling with a 500‐MHz centre frequency antenna facilitated the generation of two‐dimensional profiles and amplitude depth slices. The analyses presented here describe a processing sequence used to improve the image fidelity and the visualization of the remains of a relict archaeological excavation unit. Additionally, we employ depth slice overlay analysis for enhanced visualization of this feature. The results demonstrate a new processing methodology that helps in delineating the lateral extent of the excavation and its evolution. The results also highlight the implementation of a new visualization strategy in GPR surveys aimed towards detecting graves, burial tombs, and other rectilinear or cubic features. We conclude that even with excellent GPR imaging, there may be significant positioning errors in defining the boundaries of buried features.