3D Ground-penetrating Radar Data Research Articles

Detection of Buried Roman Wall Remains in Ground‐penetrating Radar Data using Template Matching

AbstractWhereas in the last decades the acquisition and processing of archaeological ground‐penetrating radar (GPR) data have become mature, the interpretation is still challenging. Manual delineation in three dimensions is time consuming, and often the determination of an isosurface value is not straightforward. This paper presents a method designed specifically for the extraction of buried linear features such as wall foundations, based on template matching. First, the three‐dimensional (3D) GPR data cube is synthesized into a two‐dimensional (2D) slice. To achieve this, an energy slice based on a sufficiently large time window may often be appropriate, although in this study a combination with other attributes, for example based on phase symmetry, made weak anomalies more distinct. In the next step, we compute the 2D normalized cross‐correlation of the composite 2D slice and a number of templates with dimensions similar to the walls in the data set. Of the resulting correlation matrices, the highest correlation coefficient is kept for each pixel, if it exceeds a certain threshold. In this way, wall foundations are successfully mapped, but also many false detections are produced. The latter are greatly reduced in number by using a size threshold and discarding isolated features. The remaining regions are enclosed in bounding boxes, which after vertical extrusion can be used as a simplified 3D representation of the wall structures, and for the creation of a filtered isosurface. For the evaluation of our results, a manual interpretation was used. In the 2D case (i.e. when comparing the total area of the automatically mapped structures versus the manually delineated ones), both the detection rate and the correctness were ~77%. Slightly lower rates (~71%) were obtained in the 3D case (i.e. comparing volumes). Our method was applied to the GPR survey of a Roman villa in Kent, UK. Copyright © 2016 John Wiley & Sons, Ltd.

Multichannel 3D Ground Penetrating Radar - Advances in Civil Infrastructure Scanning

The scope of this paper is to highlight improvements in Ground Penetrating Radar (GPR) as an infrastructure condition assessment tool, in particular through the use of multichannel 3D GPR. Multichannel 3D GPR is a relatively new and alternative infrastructure scanning tool which can assist geophysicists and engineers in providing 100% sub-surface coverage of an investigation area, where site access is possible. Advantages of an increased level of subsurface coverage using multichannel 3D GPR includes providing the user with improved accuracy in highlighting and quantifying regions that may require further invasive testing for future maintenance programs and also possible long-term monitoring. This paper briefly discusses current applications of standard GPR for infrastructure condition assessments and how multichannel 3D GPR can improve knowledge of the sub-surface in these application areas. Visualisations of multichannel 3D GPR data outputs with interpretations have been presented to illustrate the improved subsurface information made available from this method. The example presented is an approximately 4 m long section of multichannel 3D GPR data acquired along the surface of a reinforced concrete-lined tunnel.

Imaging of an active fault: Comparison between 3D GPR data and outcrops at the Castrovillari fault, Calabria, Italy

We have integrated and analyzed a 3D ground-penetrating radar (GPR) volume with a trenching exposure data set to evaluate the potential of these methods individually and combined for study of a fault zone. We chose a test site across a branch of the active Castrovillari fault in the Northern Calabria (Southern Italy). This tectonic structure is one of the most active in the area, and it has generated strong earthquakes in the past. Based on analysis of previously collected data, a 3D GPR survey was carried out 1.2 m from a fault outcrop. The goal was to use the GPR volume to guide and optimize the excavation of a trench and then to use the trenching data to validate the GPR volume interpretation. We used seismic interpretation software to display vertical and horizontal sections and for horizon tracking and attribute analyses. We obtained quantitative information on the geometry of structural and geologic features, such as fault strike and dip angle, defining the boundaries of different stratigraphic units. We validated our GPR data interpretation with the outcrop section and trench wall demonstrating the benefits of GPR in extensional tectonics environments and the great potential of the combined geologic and geophysical approach.

Diffraction signatures of fracture intersections

Fractured rock causes diffractions, which are often discarded as noise in ground-penetrating radar (GPR) and seismic data. Most fractures are too thin, too steep, and their displacement is too small to be imaged by reflections, and diffractions are the only detectable signal. To decipher the information about fracture geometry and distribution contained in diffractions, we compare 3D synthetic ray-Born modeling with high-density 3D GPR data and outcrop observations from the Cassis Quarry in Southern France. Our results reveal how the intersection between two fractures is the basic geologic element producing a recordable diffraction. In this new model, two intersecting fractures are represented by one finite-length line diffractor. The intersection of three fractures is a 3D cross composed of three line diffractors. Fractures extending over several meters in the outcrop display linear clusters of diffraction circles in unmigrated GPR time slices. Such large-scale fracture intersections are composed of many aligned short subwavelength line diffractors due to fracture roughness and variations of fracture opening. The shape irregularities and amplitude variations of composite diffraction signatures are a consequence of the geometry and spacing of the intersecting fractures generating them. With three simple base-type intersecting fracture models (horizontal dip, gentle dip, and steep dip), the fracture network geometry can be directly deciphered from the composite diffraction signatures visible on unmigrated time slices. The nonrandom distribution of diffractions is caused by fracture trends and patterns providing information about fracture dip, spacing, and continuity of fractured domains. With the similarity law, the diffraction phenomena observed in GPR data are very similar in character to those seen on the seismic scale with the wavelength as the scaling link. GPR data serve as a proxy to decipher seismic diffractions.

Noise Attenuation from GPR Data Using Wavelet Transform and Artificial Neural Network

In this paper, a tentative of noise attenuation from the 3D Ground Penetrating Radar (GPR) data using the wavelet transform and the Multilayer Perceptron neural network model is implanted. GPR data recorded in Algerian Sahara are filtered firstly using the discrete wavelet transform, after that a MLP machine with three layers is trained in a supervised learning mode, the input is an extracted profile from the raw 3D GPR data and the output consists of the GPR data of the same profile but after filtering. The estimated weights of connection are used to propagate the remaining non-filtered data through the implanted machine, the calculated output consists to the filtered GPR data. Comparison between the filtered data using the MLP machine and the continuous wavelet transform shows that the neural network machine can be used for S/N ratio improvement of the noisy GPR data.

Multiscale analysis of noisy 3D GPR data using the directional continuous wavelet transform

Here, we present a new technique of noise effect attenuation in the 3D ground-penetrating radar (GPR) data analysis using the 2D directional continuous wavelet transform (DCWT). The proposed technique is based on the application of a Gaussian low pass filter to the modulus of the 2D DCWT for low scales. After application of the low pass filter, maxima of the CWT are mapped for all range of scales. Application to a noisy GPR data shows that the proposed idea can improve GPR data analysis by the continuous wavelet transform.

2D and 3D imaging of a buried prehistoric canoe using GPR attributes: a case study

ABSTRACTWe apply Ground Penetrating Radar (GPR) to detect a prehistoric canoe at the Maoshan site, Zhejiang Province, China. A complex attribute analysis of the GPR data allows enhancing the precision in target detection and provides more details about the canoe and the burial environment. The burial depth of the bottom interface of the prehistoric canoe is detected and the integrity of the whole canoe is assessed through a GPR survey. Difficulties in the application of dense sampling of 2D and pseudo 3D GPR data originate from micro‐topographical disturbance that specifically affects the pseudo 3D investigation results obtained from high‐frequency antennas. Data processing and advanced imaging techniques can only remove part of such effects. The research demonstrates that GPR can successfully image wooden cultural relics buried in the shallow subsurface with ultra, high‐density trace spacing and high‐frequency antennas even in totally saturated clay‐rich soils based on 2D profiles and pseudo 3D methodologies, characterized by tight (cm) cross‐line/in‐line spacing.

Architectural characterization of a delta-front reservoir analogue combining Ground Penetrating Radar and Electrical Resistivity Tomography: Roda Sandstone (Lower Eocene, Graus-Tremp basin, Spain)

Three-dimensional reconstruction of reservoir analogues can be improved combining data from different geophysical methods. Ground Penetrating Radar (GPR) and Electrical Resistivity Tomography (ERT) data are valuable tools, since they provide subsurface information from internal architecture and facies distribution of sedimentary rock bodies, enabling the upgrading of depositional models and heterogeneity reconstruction. The Lower Eocene Roda Sandstone is a well-known deltaic complex widely studied as a reservoir analogue that displays a series of sandstone wedges with a general NE to SW progradational trend. To provide a better understanding of internal heterogeneity of a 10m-thick progradational delta-front sandstone unit, 3D GPR data were acquired. In addition, common midpoints (CMP) to measure the sandstone subsoil velocity, test profiles with different frequency antennas (25, 50 and 100MHz) and topographic data for subsequent correction in the geophysical data were also obtained. Three ERT profiles were also acquired to further constrain GPR analysis. These geophysical results illustrate the geometry of reservoir analogue heterogeneities both depositional and diagenetic in nature, improving and complementing previous outcrop-derived data. GPR interpretation using radar stratigraphy principles and attributes analysis provided: 1) tridimensional geometry of major stratigraphic surfaces that define four units in the GPR Prism, 2) image the internal architecture of the units and their statistical study of azimuth and dips, useful for a quick determination of paleocurrent directions. These results were used to define the depositional architecture of the progradational sandbody that shows an arrangement in very-high-frequency sequences characterized by clockwise paleocurrent variations and decrease of the sedimentary flow, similar to those observed at a greater scale in the same system. This high-frequency sequential arrangement has been attributed to the autocyclic dynamics of a supply-dominated delta-front where fluvial and tidal currents are in competition. The resistivity models enhanced the viewing of reservoir quality associated with cement distribution caused by depositional and early diagenetic processes related to the development of transgressive and regressive systems tracts in high-frequency sequences.

Multiscale analysis of 3D GPR data using the 2D directional continuous wavelet transform (DCWT)

The main goal of this paper is to extract land topographical information by analyzing the 3D ground-penetrating radar (GPR) data by the 2D directional continuous wavelet transform method. The first objective is to identify direction of obstacles, and the second is to characterize each one with a special roughness coefficient. Application on real data shows that the proposed tool can enhance the GPR data interpretation.

Imaging karstic cavities in transparent 3D volume of the GPR data set in Akkopru dam, Mugla, Turkey

The use of geophysical methods, particularly ground-penetrating radar (GPR), in the environmental sites has taken very important approximation in locating potential environmental hazards such as landslides and cavities. Generally, slices of a three-dimensional (3D) GPR data set can be used to monitor changes in the surface locations and depth of the environmental hazards through the iterative slices. The aim of this study is to introduce volume imaging of subsurface cavities with a transparent half bird's-eye view within a 3D data set. Transparency is obtained by a constructed opaque function of the amplitude–colour range. The half bird's-eye view is achieved by observing angles of the x-, y- and z-directions to the 3D block. A GPR study was conducted at a reservoir area (696 m × 100 m) in Akkopru dam and hydroelectric power station in Mugla (Turkey) to determine the existence and abundance of karstic cavities. The study revealed the presence of 283 cavities of various size and depth. Borehole data encouraged the results with given depths.

Biographies of the guest editors

Laura Valentina Socco took her PhD in environmental geo-engineering in 1996 at Politecnico di Torino, where she is presently Assistant Professor of Applied Geophysics. Her main research activity is focused on characterization techniques based on seismic surface wave analysis and geophysical data integration. Xander’s affinity with near-surface geophysical methods is a result of his thesis work on 3D ground-penetrating radar data imaging, in the Mining Department at Delft University, the Netherlands. His interest in surface waves started during his PhD in the Applied Mathematics Dept, also at Delft University, where from 2000-2005, he worked on the development of a method to predict scattered surface waves in exploration seismic data.

Integrated data analysis at an archaeological site: A case study using 3D GPR, magnetic, and high-resolution topographic data

We have collected magnetic, 3D ground-penetrating-radar (GPR), and topographic data at an archaeological site within the Palace Garden of Paretz, Germany. The survey site covers an area of approximately [Formula: see text] across a hill structure (dips of up to 15°) that is partly covered by trees. The primary goal of this study was to detect and locate the remains of ancient architectural elements, which, from historical records, were expected to be buried in the subsurface at this site. To acquire our geophysical data, we used a recently developed surveying approach that combines the magnetic and GPR instrument with a tracking total station (TTS). Besides efficient data acquisition, this approach provides positional information at an accuracy within the centimeter range. At the Paretz field site, this information was critical for processing and analyzing our geophysical data (in particular, GPR data) and enabled us to generate a high-resolution digital terrain model (DTM) of the surveyed area. Integrated analysis and interpretation based on composite images of the magnetic, 3D GPR, and high-resolution DTM data as well as selected attributes derived from these data sets allowed us to outline the remains of an artificial grotto and temple. Our work illustrates the benefit of using multiple surveying technologies, analyzing and interpreting the resulting data in an integrated fashion. It further demonstrates how modern surveying solutions allow for efficient, accurate data acquisition even in difficult terrain.

Improving the interpretability of 3D GPR data using target–specific attributes: application to tomb detection

Three–dimensional (3D) ground–penetrating radar (GPR) represents an efficient high–resolution geophysical surveying method allowing to explore archaeological sites in a non–destructive manner. To effectively analyze large 3D GPR data sets, their combination with modern visualization techniques (e.g., 3D isoamplitude displays) has been acknowledged to facilitate interpretation beyond classical time–slice analysis. In this study, we focus on the application of data attributes (namely energy, coherency, and similarity), originally developed for petroleum reservoir related problems addressed by reflection seismology, to emphasize temporal and spatial variations within GPR data cubes. Based on two case studies, we illustrate the potential of such attribute based analyses towards a more comprehensive 3D GPR data interpretation. The main goal of both case studies was to localize and potentially characterize tombs inside medieval chapels situated in the state of Brandenburg, Germany. By comparing the calculated data attributes to the conventionally processed data cubes, we demonstrate the superior interpretability of the coherency and the similarity attribute for target identification and characterization.

Full-Resolution 3D Radar Stratigraphy of Complex Oolitic Sedimentary Architecture: Miami Limestone, Florida, U.S.A.

Abstract At the field scale (m's to 100's m) it is often difficult to define the true three-dimensional form of depositional units or facies within sediments or sedimentary rocks, due to inherent spatial bias in traditional data sources such as outcrops and cores. Full-resolution 3D ground-penetrating radar (GPR) surveying has the potential to overcome this problem. Southwest of Miami, paleotopography on the upper Pleistocene Miami Limestone Formation indicates the former presence of a large oolitic barrier bar, backed by extensive ooid shoals. Data from three closely spaced cores and two large outcrops on the barrier bar allow initial subdivision of the Miami Limestone in to a lower mottled oolitic facies and an upper bedded oolitic facies. However, detailed correlation between the cores and outcrops is problematic due to spatial aliasing. A 250 MHz full-resolution 3D GPR data set was collected over an area of 45 m × 24 m that encompassed the location of the three cores. Depth of radar wave penetration was around 6 m, and both horizontal and vertical resolution was approximately 0.1 m for the 6500 m3 rock volume, such that the upper bedded oolitic facies was imaged without spatial aliasing. Radar surfaces and intervening radar facies define two main radar package types: low-angle sheet and sigmoidal. The low-angle sheet radar packages are interpreted as oolitic shoal deposits consisting of tidal bars covered by ebb-orientated, straight-crested sand waves. The overlying sigmoidal radar packages record barrier-bar formation through alongshore progradation. Comparable sedimentary successions from both Pleistocene and modern deposits in the Bahamas suggest that the three evolutionary stages recorded for the Miami Limestone are typical of many transgressive ooid shoal complexes.

What can GPR tell us about cryoconite holes? 3D FDTD modeling, excavation and field GPR data

Cryoconite holes form on ice due to enhanced ablation around particles deposited on the surface, and are present in the ablation area of glaciers worldwide. Here we investigate the use of Ground Penetrating Radar (GPR) as a non-destructive method to monitor and map cryoconite holes. We compare GPR data obtained from the Jutulsessen blue ice area in Dronning Maud Land, Antarctica, with modeled GPR data. The modeled GPR response to cryoconite holes is numerically calculated by solving Maxwell's equations with a 3D Finite-Difference Time-Domain (FDTD) scheme. The model includes a realistic shielded bowtie antenna and dimensions and constituent parameters of cryoconite holes excavated in the field. We have performed what-if scenarios with controlled variation of single parameters. We show that GPR can be used to determine the horizontal extent, depth and whether a cryoconite hole is frozen or contains liquid water, information unavailable from visual surface inspection. The cryoconite thickness can, for completely frozen holes, be determined to within a 1/4 of the GPR center frequency wavelength. The exact water content is not readily extractable because the GPR response is influenced by many other factors such as: cryoconite thickness, shape and roughness, as well as antenna ground coupling.

Visualization of active faults using geometric attributes of 3D GPR data: An example from the Alpine Fault Zone, New Zealand

Three-dimensional ground-penetrating radar (GPR) data are routinely acquired for diverse geologic, hydrogeologic, archeological, and civil engineering purposes. Interpretations of these data are invariably based on subjective analyses of reflection patterns. Such analyses are heavily dependent on interpreter expertise and experience. Using data acquired across gravel units overlying the Alpine Fault Zone in New Zealand, we demonstrate the utility of various geometric attributes in reducing the subjectivity of 3D GPR data analysis. We use a coherence-based technique to compute the coherency, azimuth, and dip attributes and a gray-level co-occurrence matrix (GLCM) method to compute the texture-based energy, entropy, homogeneity, and contrast attributes. A selection of the GPR attribute volumes allows us to highlightkey aspects of the fault zone and observe important features not apparent in the standard images. This selection also provides information that improves our understanding of gravel deposition and tectonic structures at the study site. A new depositional/structural model largely based on the results of our analysis of GPR attributes includes four distinct gravel units deposited in three phases and a well-defined fault trace. This fault trace coincides with a zone of stratal disruption and shearing bound on one side by upward-tilted to synclinally folded stratified gravels and on the other side by moderately dipping stratified alluvial-fan gravels that could have been affected by lateral fault drag. When used in tandem, the coherence- and texture-based attribute volumes can significantly improve the efficiency and quality of 3D GPR interpretation, especially for complex data collected across active fault zones.

Semi‐automatic feature extraction from GPR data for archaeology

ABSTRACTA new method and working tool to interpret 3D ground‐penetrating radar data has been developed. Algorithms for feature extraction based on three‐dimensional regression techniques are developed to generate 3D models of the archaeological reflection features in the subsurface.After several attempts to reasonably visualize the information of a 3D ground‐penetrating radar (GPR) data cube by manual and automatic techniques, we propose a new approach: the visualization and interpretation steps are combined in an efficient and objective process in order to extract the geometrical information of the mapped archaeological features. Each object is modelled by six independent planes that form the structure that should be extracted. This is an appropriate approximation for walls, floors, stones as well as for pipes. To speed up calculation and because routines operating on the complete data set were not successful so far, the system utilizes a start line which is defined by the user. After 3D pre‐filtering with a sequence of median – dilation – erosion, followed by a 3D gradient filter, planes are matched to the borders of the anomaly by using a least‐median‐of‐squares algorithm. Finally, the found planes are refined by an analytical regression and the final box displayed in a 3D GIS software. The small computing time for a single object makes this method feasible for routine application of feature extraction from a 3D GPR data cube. The presented technique allows visualizing and extracting the imaged archaeological features rapidly and into an easy understandable representation.

Spectral decomposition of 3D ground-penetrating radar data from an alluvial environment

This study uses spectral decomposition of GPR data to recognize tuning in a wedge of unconsolidated alluvial sediments. It also compares the spectral decomposition analysis to the results of more traditional attributes such as instantaneous frequency. The study area is on the northern bank of the Canadian River, south of Norman, Oklahoma, USA. (Figure 1). The alluvium ranges in thickness from approximately 10 to 15 m and is mostly sand, with varying amounts of silt and clay and rare occurrences of gravel. Underlying the alluvium is the Hennessey Shale.

Sensor Fusion for the Detection of Landmines

The detection of buried anti-personnel mines (APMs) is widely considered as a problem which may only be solved with a combination of two or more complementary sensors. We present processing and fusion results obtained from a multisensor data set, acquired with a pulse induction metal detector (MD), a pulsed ultra wide band ground penetrating radar (GPR) and a 3–5μ thermal infra-red (IR) camera. Various types of soils, clutter objects and burial conditions were recorded. Anti-personnel mines included minimum metal mines as well as mines with a significant metal content. We use a special projection to map a 3D GPR data cube, with time or depth as vertical co-ordinate, into a horizontal plane view 2D image. Object contours are then derived, based on an edge extraction method, followed by an automatic detection of circular shapes with a Hough-transform. In the association step, the stand-off IR image, the metal detector and GPR images and related detections are mapped onto a common cartesian grid on the ground surface. Detection results are fused on a decision level, using a Bayesian approach. Our results indicate that the GPR performance approximately matches that of the metal detector. With both sensors all metallic mines and around 60% of the minimum metal mines were detected. In the case of two false alarms per square meter combined detection probability clearly exceeds single sensor performance.

New brochure highlights diversity of Albright & Wilson

Three–dimensional (3D) ground–penetrating radar (GPR) represents an efficient high–resolution geophysical surveying method allowing to explore archaeological sites in a non–destructive manner. To effectively analyze large 3D GPR data sets, their combination with modern visualization techniques (e.g., 3D isoamplitude displays) has been acknowledged to facilitate interpretation beyond classical time–slice analysis. In this study, we focus on the application of data attributes (namely energy, coherency, and similarity), originally developed for petroleum reservoir related problems addressed by reflection seismology, to emphasize temporal and spatial variations within GPR data cubes. Based on two case studies, we illustrate the potential of such attribute based analyses towards a more comprehensive 3D GPR data interpretation.The main goal of both case studies was to localize and potentially characterize tombs inside medieval chapels situated in the state of Brandenburg, Germany. By comparing the calculated data attributes to the conventionally processed data cubes, we demonstrate the superior interpretability of the coherency and the similarity attribute for target identification and characterization.