Three-dimensional Ground-penetrating Radar Research Articles

The influence of interlayer bonding conditions on the propagation laws of reflective cracks in semi-rigid base pavement based on the DEM and GPR

Reflection cracks represent the predominant internal distress in semi-rigid base asphalt pavements and often accompanied by additional secondary issues that detrimentally affect the pavement structural bearing capacity. The propagation and deterioration of reflective cracks are affected by many factors, for example the interlayer bonding conditions. It is crucial to identify the propagation and deterioration of reflective cracks to prolong the life of pavement structure. The purpose of this article is to explore the influence of interlayer bonding conditions on the propagation laws of reflective cracks. This work employs the discrete element method(DEM) to explore the propagation patterns of reflection cracks accompanied different bonding conditions between the surface layer and base layer subjected to moving load. The results of discrete element simulation reveal a three-phase progression laws when the reflective cracks expand from the base layer to the pavement surface，involving rapid propagation, stable propagation, and failure process. With the interlayer bond strength decreases, expansion of reflective crack is significantly accelerated. Meanwhile, once the critical bond strength threshold is reached, complete debonding will occur between the surface and base layers. The results are ultimately verified through the core sampling from practical engineering using three-dimensional (3D) ground penetrating radar(GPR) detection, along with the radar image recognition of reflective cracks and poor interlayer bonding. The results will serve as a foundation for researching the mechanism and treatment measures for two internal pavement distress, namely reflection cracks and poor interlayer bonding.

Three Dimension Ground Penetrating Radar and Data Processing for Pavement Collapse Detection

Ground-penetrating radar (GPR) is a non-destructive testing exploration technology with the advantages of high efficiency and high resolution, and is widely used in roadbed detection and disease detection. With the increasing demand for road detection, traditional handheld two-dimensional GPR detection instruments can no longer meet the normal urban road detection requirements. Therefore, new efficient vehicle-mounted three-dimensional GPR systems have emerged and been put into use. This article mainly introduces some three-dimensional GPR instruments and systems that have been put into use, analyzes their composition structure, usage characteristics, compares their similarities and differences, and data processing methods, which can provide a certain reference for the forward simulation and data processing of three-dimensional GPR.

Clustering of Asphalt Pavement Maintenance Sections Based on 3D Ground-Penetrating Radar and Principal Component Techniques

Asphalt pavement maintenance section classification is an important prerequisite for accurately determining asphalt pavement maintenance needs and formulating accurate maintenance plans. This paper introduces the three-dimensional (3D) ground-penetrating radar (GPR) pavement internal crack rate index on the basis of an original road surface performance data matrix, and the dimensionality of the road section classification data matrix was reduced through the principal component technique. An analysis of variance was used to compare the significance of the differences in the results for road section classification using different clustering methods and different clustering data and to investigate the influence of the clustering method, principal component technique and crack rate index on the maintenance road section classification results. The results showed that the principal component technique could reduce the dimensionality of the data matrix by 33% and retain more than 84% of the information. There was a genetic relationship between the clustering data and the technical characteristics of the classified sub-sections, and the internal crack rate was important for the characterisation of internal defects in asphalt pavement sub-sections and the determination of maintenance needs. The results of section classification varied considerably between clustering methods, and the choice of clustering method had a relationship to the pavement maintenance objectives. The dynamic clustering method combined with principal component analysis could significantly improve the significance of the differences in the clustering results, effectively improving the division of maintenance sections.

Intensity Normalisation of GPR C-Scans

The three-dimensional (3D) ground-penetrating radar (GPR) has been widely applied in subsurface surveys and imaging, and the quality of the resulting C-scan images is determined by the spatial resolution and visualisation contrast. Previous studies have standardised the suitable spatial resolution of GPR C-scans; however, their measurement normalisation remains arbitrary. Human bias is inevitable in C-scan interpretation because different visualisation algorithms lead to different interpretation results. Therefore, an objective scheme for mapping GPR signals after standard processing to the visualisation contrast should be established. Focusing on two typical scenarios, a reinforced concrete structure and an urban underground, this study illustrated that the essential parameters were greyscale thresholding and transformation mapping. By quantifying the normalisation performance with the integration of image segmentation and structural similarity index measure, a greyscale threshold was developed in which the normalised standard deviation of the unit intensity of any surveyed object was two. A transformation function named “bipolar” was also shown to balance the maintenance of real reflections at the target objects. By providing academia/industry with an object-based approach, this study contributes to solving the final unresolved issue of 3D GPR imaging (i.e., image contrast) to better eliminate the interfering noise and better mitigate human bias for any one-off/touch-based imaging and temporal change detection.

Crack Unet: Crack Recognition Algorithm Based on Three-Dimensional Ground Penetrating Radar Images

Three-dimensional (3D) ground-penetrating radar is an effective method for detecting internal crack damage in pavement structures. Inefficient manual interpretation of radar images and high personnel requirements have substantially restrained the generalization of 3D ground-penetrating radar. An improved Crack Unet model based on the Unet semantic segmentation model is proposed herein for 3D ground-penetrating radar crack image processing. The experiment showed that the MPA, MioU, and accuracy of the model were improved, and it displayed better capacity in the radar image crack segmentation task than current mainstream algorithms do, such as deepLabv3, PSPNet, and Unet. In the test dataset without cracks, Crack Unet is on the same level as deepLabv3 and PSPNet, which can meet engineering requirements and display a significant improvement compared with Unet. According to the ablation experiment, the MPA and MioU of Unet configured with PMDA, MC-FS, and RS modules were larger than those of Unet configured with one or two modules. The PMDA module adopted by the Crack Unet model showed a higher MPA and MioU than the SE module and the CBAM module did, respectively. The results show that the Crack Unet model has a better segmentation ability than the current mainstream algorithms do in the task of the crack segmentation of radar images, and the performance of crack segmentation is significantly improved compared with the Unet model. The Crack Unet model has excellent engineering application value in the task of the crack segmentation of radar images.

Low Compaction Level Detection of Newly Constructed Asphalt Pavement Based on Regional Index.

In order to improve the prediction accuracy regarding low compaction level of asphalt pavement, this paper carries out indoor tests to detect the voids and dielectric constants of AC-13, AC-16 and AC-25 asphalt mixtures, obtaining their relationship equations via linear fitting and determining the dielectric constant judgment threshold of low compaction level segregation risk points . Based on the common mid-point method, three-dimensional ground-penetrating radar is used to obtain the dielectric constant of the physical engineering test section. The researcher can draw the distribution map of the low compaction level segregation risk area according to the judgment threshold of the rough segregation risk points; divide the connected risk areas; determine the regional convex hull; and calculate the regional indicators such as the regional area, the ratio of the convex risk points and the mean value of the regional dielectric constant. The response surface analysis method is used to acquire the model of risk area index and core void ratio. The model is employed to predict and verify the core void ratio in the risk area of the road section and verify the accuracy of the model. The results show that the error range between the predicted voids and the measured voids is −0.4%~+0.4%, and the mean absolute value of the error is 0.25%. Compared with the mean measured voids of 6.63%, the relative error is 3.77%, indicating that the model can accurately predict the regional low compaction level segregation degree.

Fast Segmentation and Dynamic Monitoring of Time-Lapse 3D GPR Data Based on U-Net

As the amount of ground-penetrating radar (GPR) data increases significantly with the high demands of nondestructive detection methods under urban roads, a method suitable for time-lapse data dynamic monitoring should be developed to quickly identify targets on GPR profiles and compare time-lapse datasets. This study conducted a field experiment aiming to monitor one backfill pit using three-dimensional GPR (3D GPR), and the time-lapse data collected over four months were used to train U-Net, a fast neural network based on convolutional neural networks (CNNs). Consequently, a trained network model that could effectively segment the backfill pit from inline profiles was obtained, whose Intersection over Union (IoU) was 0.83 on the test dataset. Moreover, segmentation masks were compared, demonstrating that a change in the southwest side of the backfill pit may exist. The results demonstrate the potential of machine learning algorithms in time-lapse 3D GPR data segmentation and dynamic monitoring.

Pavement Distress Detection Using Three-Dimension Ground Penetrating Radar and Deep Learning

Three-dimensional ground penetrating radar (3D GPR) is a non-destructive examination technology for pavement distress detection, for which its horizontal plane images provide a unique perspective for the task. However, a 3D GPR collects thousands of horizontal plane images per kilometer of the investigated pavement. The existing detection methods using GPR images are time-consuming and risky for subjective judgment. To solve the problem, this study used deep learning methods and 3D GPR horizontal plane images to detect pavement structural distress, including cracks, repairs, voids, poor interlayer bonding, and mixture segregation. In this study, two deep learning methods, called CP-YOLOX and SViT, were used to achieve the aim. A dataset for anomalous waveform localization (3688 images) was first created by pre-processing 3D-GPR horizontal plane images. A CP-YOLOX model was then trained to localize anomalous waveforms. Five SViT models with different numbers of encoders were adopted to perform the classification of anomalous waveforms using the localization results from the CP-YOLOX model. The numerical experiment results showed that 3D GPR horizontal plane images have the potential to be an assistant for pavement structural distress detection. The CP-YOLOX model achieved 87.71% precision, 80.64% mAP, and 33.57 sheets/s detection speed in locating anomalous waveforms. The optimal SViT achieved 63.63%, 68.12%, and 75.57% classification accuracies for the 5-category, 4-category, and 3-category datasets, respectively. The proposed models outperformed other deep learning methods on distress detection using 3D GPR horizontal plane images. In the future, more radar images should be collected to improve the accuracy of SViT.

Research on the Quality of Asphalt Pavement Construction Based on Nondestructive Testing Technology

In order to better evaluate the construction quality of asphalt pavement, nondestructive testing techniques are used to inspect newly paved asphalt mixture pavement. The proposed system for the evaluation of asphalt pavement construction quality uses three-dimensional ground-penetrating radar (GPR) and a non-nuclear density gauge. The GPR and the non-nuclear density gauge test results were used to establish a dielectric constant–porosity model by fitting. This approach can more accurately determine the dielectric constant selection scheme of the GPR based on the average value of every 10 dielectric constant data points in the length direction of the radar antenna and every three data channels in the width direction. The GPR collected the dielectric constants of the road surface based on the total reflection method and used the average value of the local dielectric constant to evaluate the construction quality of the road. The non-nuclear density gauge used the local porosity to assess the construction quality of the road. It is recommended that the two testing schemes described above be used to evaluate the quality of asphalt pavement construction. They can provide theoretical guidance for future applications in practical processes.

Use of NDT systems to investigate pavement reconstruction needs and improve maintenance treatment decision-making

ABSTRACT The objective is to improve the accuracy of the test results and establish a maintenance treatment decision-making system using the non-destructive testing (NDT) technology. Firstly, the evaluation unit length was theoretically determined and statistically verified. Secondly, the test frequency of falling weight deflector (FWD) test was statistically analysed and the deflection acceptance criterionwas obtained based on the probability density distribution curve. Meanwhile, the relationship between the FWD deflection and the pavement internal damage obtained by drilling was analysed. Subsequently, the test parameters of the three-dimensional ground-penetrating radar (3D GPR) were optimised by using the non-reference image quality assessing technology. Followed, pavement internal damage index of GPR was proposed for locating the defect structural layer. Lastly, the decision-making systemwas established. Results showed that the suitable evaluation unit length is 20 m, the deflection test frequency is no more than 10 m/point, the deflection accept criterion is 29.5 (0.01 mm) with 95% reliability and the optimal detection parameters of GPR is 1 cm, 10 µs and 25 ns respectively. The established decision-making system fully considering the surface and the interior conditions as well as the functional and structural conditions can effectively guide the maintenance plan formulation for similar reconstruction and expansion projects.

Identifying airport runway pavement diseases using complex signal analysis in GPR post-processing

Non-destructive detection of pavement diseases on airport runways is of great importance for airport maintenance and ensuring the normal use of airport runways. In this study, a full coverage scan of an airport pavement was conducted using the MALA three-dimensional ground penetrating radar (GPR). Based on complex signal analysis technology, the Hilbert transform were used to comprehensively interpret the anomalous features of the scanned images at specific disease location. The theoretical void volume of pavement diseases was calculated using MATLAB image processing and Origin software. The analysis showed that the main airport pavement diseases were loose layers and small holes, which were more comprehensively revealed by three-dimensional radar images. The three instantaneous parameter spectrums were used to verify and optimize the initial results obtained with the Reflexw post-processing software. The theoretical void volume was calculated using the pixel equivalent method. The void value of typical diseases was 49.38 m3, which was approximately 2.97 times of the calculated value with the rSlicer post-processing technique. This proposed technique can provide a reference for the exact recognition of disease types and the corresponding calculation of grouting reinforcements while overcoming the shortcomings of the single parameter method with multi-parameter analysis.

Mapping of Agricultural Subsurface Drainage Systems Using a Frequency-Domain Ground Penetrating Radar and Evaluating Its Performance Using a Single-Frequency Multi-Receiver Electromagnetic Induction Instrument.

Subsurface drainage systems are commonly used to remove surplus water from the soil profile of a poorly drained farmland. Traditional methods for drainage mapping involve the use of tile probes and trenching equipment that are time-consuming, labor-intensive, and invasive, thereby entailing an inherent risk of damaging the drainpipes. Effective and efficient methods are needed in order to map the buried drain lines: (1) to comprehend the processes of leaching and offsite release of nutrients and pesticides and (2) for the installation of a new set of drain lines between the old ones to enhance the soil water removal. Non-invasive geophysical soil sensors provide a potential alternative solution. Previous research has mainly showcased the use of time-domain ground penetrating radar, with variable success, depending on local soil and hydrological conditions and the central frequency of the specific equipment used. The objectives of this study were: (1) to test the use of a stepped-frequency continuous wave three-dimensional ground penetrating radar (3D-GPR) with a wide antenna array for subsurface drainage mapping and (2) to evaluate its performance with the use of a single-frequency multi-receiver electromagnetic induction (EMI) sensor in-combination. This sensor combination was evaluated on twelve different study sites with various soil types with textures ranging from sand to clay till. While the 3D-GPR showed a high success rate in finding the drainpipes at five sites (sandy, sandy loam, loamy sand, and organic topsoils), the results at the other seven sites were less successful due to the limited penetration depth of the 3D-GPR signal. The results suggest that the electrical conductivity estimates produced by the inversion of apparent electrical conductivity data measured by the EMI sensor could be a useful proxy for explaining the success achieved by the 3D-GPR in finding the drain lines.

Three-dimensional convolutional neural network–based underground object classification using three-dimensional ground penetrating radar data

Ground-penetrating radar is a typical sensor system for analyzing underground facilities such as pipelines and rebars. The technique also can be used to detect an underground cavity, which is a potential sign of urban sinkholes. Multichannel ground-penetrating radar devices are widely used to detect underground cavities thanks to the capacity of informative three-dimensional data. Nevertheless, the three-dimensional ground-penetrating radar data interpretation is unclear and complicated when recognizing underground cavities because similar ground-penetrating radar data reflected from different underground objects are often mixed with the cavities. As it is prevalently known that the deep learning algorithm-based techniques are powerful at image classification, deep learning-based techniques for underground object detection techniques using two-dimensional GPR (ground-penetrating radar) radargrams have been researched upon in recent years. However, spatial information of underground objects can be characterized better in three-dimensional ground-penetrating radar voxel data than in two-dimensional ground-penetrating radar images. Therefore, in this study, a novel underground object classification technique is proposed by applying deep three-dimensional convolutional neural network on three-dimensional ground-penetrating radar data. First, a deep convolutional neural network architecture was developed using three-dimensional convolutional networks for recognizing spatial underground objects such as, pipe, cavity, manhole, and subsoil. The framework of applying the three-dimensional convolutional neural network into three-dimensional ground-penetrating radar data was then proposed and experimentally validated using real three-dimensional ground-penetrating radar data. In order to do that, three-dimensional ground-penetrating radar block data were used to train the developed three-dimensional convolutional neural network and to classify unclassified three-dimensional ground-penetrating radar data collected from urban roads in Seoul, South Korea. The validation results revealed that four underground objects (pipe, cavity, manhole, and subsoil) are successfully classified, and the average classification accuracy was 97%. In addition, a false alarm was rarely indicated.

Planform and stratigraphic signature of proximal braided streams: remote-sensing and ground-penetrating-radar analysis of the Kicking Horse River, Canadian Rocky Mountains

ABSTRACTBraided rivers have accumulated a dominant fraction of the terrestrial sedimentary record, and yet their morphodynamics in proximal intermountain reaches are still not fully documented—a shortcoming that hampers a full understanding of sediment fluxes and stratigraphic preservation in proximal-basin tracts. Located in the eastern Canadian Cordillera near the continental divide, the Kicking Horse River is an iconic stream that has served as a model for proximal-braided rivers since the 1970s. Legacy work on the river was based solely on ground observations of small, in-channel bars; here we integrate field data at the scale of individual bars to the entire channel belt with time-lapse remote sensing and ground-penetrating-radar (GPR) imaging, in order to produce a more sophisticated morphodynamic model for the river.Cyclical discharge fluctuations related to both diurnal and seasonal variations in melt-water influx control the planform evolution and corresponding stratigraphic signature of trunk channels, intermittently active anabranch channels, and both bank-attached and mid-channel bars. Three-dimensional GPR fence diagrams of compound-bar complexes are built based on the identification of distinct radar facies related to: i) accretion and migration of unit bars, ii) both downstream and lateral outbuilding of bar-slip foresets; iii) buildup of bedload sheets, iv) channel avulsion, and v) accretion of mounded bars around logs or outsized clasts. Trends observed downstream-ward include decreases in gradient and grain size decreases, trunk-channel shrinkage, intensified avulsion (with increase in abundance for anabranch channels), and a shift from high-relief to low-relief bar topography. The integration of ground sedimentology, time-lapse remote sensing, and GPR imaging demonstrates that proximal-braided streams such as the Kicking Horse River can be critically compared to larger systems located farther away from their source uplands despite obvious scale differences.

Three-Dimensional Reverse Time Migration of Ground-Penetrating Radar Signals

Three-dimensional (3D) ground-penetrating radar (GPR) systems and 3D seismic imaging techniques have been developing fast and evolving rapidly in the last decade. Ray-based migration methods have been successfully applied to processing 3D GPR signals based on the similarity between electromagnetic and seismic waves. However, reverse time migration (RTM) of 3D GPR signals has not been well studied in the past. In this paper, we present a 3D RTM based on Maxwell’s equations for 3D GPR surveys. Migration recovers the true subsurface structure from a distorted and unfocused time profile and suppresses common electromagnetic clutter and noise. RTM based on Maxwell’s equations can consider conductivity directly and compensate for the attenuation within a high-conductivity zone. Compared with 2D RTM, 3D RTM back-propagates both in-line and cross-line signals simultaneously and can include complex 3D permittivity and conductivity models. We have integrated a parallel finite-difference time-domain (FDTD) algorithm based on a hybrid MPI and OpenMP scheme to reduce the computational cost of 3D problems. The 3D RTM experiments on an anomaly of “EM” shape and a realistic sand dune model demonstrate the effective recovery of 3D subsurface structures.

The Messapic Site of Muro Leccese: New Results from Integrated Geophysical and Archaeological Surveys

The regular application of geophysical survey techniques to evaluate archaeological sites is well established as a method for locating, defining, and mapping buried archaeological materials. However, it is not always feasible to apply a range of different methods over a particular site or landscape due to constraints in time or funding. This paper addresses the integrated application of three geophysical survey methods over an important archaeological site located in south Italy. In particular, it is focused on the results achieved from a past geophysical survey and the ongoing excavations performed by archaeologists in the site of Muro Leccese. Muro Leccese (Lecce) is one of the most important Messapian archaeological sites in southern Italy. The archaeological interest of the site was generated since the discovery of the remains of Messapian walls (late 4th–3rd centuries BC). With the aim of widening the archaeological knowledge of the Messapian settlement, several integrated methods, including magnetometry, ground-penetrating radar, and electrical resistivity tomography were used on site to fulfill a number of different research objectives. Since the most important targets were expected to be located at shallow soil depth, a three-dimensional (3D) ground-penetrating radar (GPR) survey was carried out in two zones, which were labeled respectively as zone 1 and zone 2, and were both quite close to the archaeological excavations. The GPR investigations were integrated with a 3D electrical resistivity tomography (ERT) survey in zone 1 and with a magnetometric, in gradiometry configuration survey in zone 2. The integration of several techniques allowed mapping the structural remains of this area and leading the excavation project. The geophysical results show a good correspondence with the archaeological features that were found after the excavation. Current work on the geophysical survey data using different codes for the processing of the data and merging different datasets using a Geographic Information System allowed achieving a user-friendly visualization that was presented to the archaeologists.

An integrated high-resolution geophysical and geologic visualization of a Lake Bonneville shoreline deposit (Utah, USA)

Pilot Valley, located in the eastern Basin and Range, Western Utah, USA, contains numerous shorelines and depositional remnants of Late Pleistocene Lake Bonneville. These remnants present excellent ground-penetrating radar (GPR) targets due to their coherent stratification, low-clay, low-salinity, and low moisture content. Three-dimensional GPR imaging can resolve fine-scale stratigraphy of these deposits down to a few centimeters, and when combined with detailed outcrop characterization, it provides an in-depth look at the architecture of these deposits. On the western side of Pilot Valley, a well-preserved late Pleistocene gravel bar records shoreline depositional processes associated with the Provo (or just post-Provo) shoreline period. GPR data, measured stratigraphic sections, cores, paleontological sampling for paleoecology and radiocarbon dating, and mineralogical analysis permit a detailed reconstruction of the depositional environment of this well-exposed prograding gravel bar. Contrary to other described Bonneville shoreline deposits, calibrated radiocarbon ages ranging from 16.5 to 14.3 (ka, BP) indicate that the bar was stable and active during an overall regressive stage of the lake, as it dropped from the Provo shoreline (or just post-Provo level). Our study provides a model for an ancient pluvial lakeshore depositional environment in the Basin and Range province and suggests that stable, progradational bedforms common to the various stages of Lake Bonneville are likely not all associated with periods of shoreline stability, as is commonly assumed. The high-resolution GPR visualization demonstrates the high degree of compartmentalization possible for a potential subsurface reservoir target based on ancient shoreline sedimentary facies.

Deep learning-based automated underground cavity detection using three-dimensional ground penetrating radar

Three-dimensional ground penetrating radar data are often ambiguous and complex to interpret when attempting to detect only underground cavities because ground penetrating radar reflections from various underground objects can appear like those from cavities. In this study, we tackle the issue of ambiguity by proposing a system based on deep convolutional neural networks, which is capable of autonomous underground cavity detection beneath urban roads using three-dimensional ground penetrating radar data. First, a basis pursuit-based background filtering algorithm is developed to enhance the visibility of underground objects. The deep convolutional neural network is then established and applied to automatically classify underground objects using the filtered three-dimensional ground penetrating radar data as represented by three types of images: A-, B-, and C-scans. In this study, we utilize a novel two-dimensional grid image consisting of several B- and C-scan images. Cavity, pipe, manhole, and intact features extracted from in situ three-dimensional ground penetrating radar data are used to train the convolutional neural network. The proposed technique is experimentally validated using real three-dimensional ground penetrating radar data obtained from urban roads in Seoul, South Korea.

Dissecting and Interpreting a Three-Dimensional Ground-Penetrating Radar Dataset: An Example from Northern Australia.

A robust 3-D GPR dataset provides interpreters with a variety of methods for extracting important information at buried archaeological sites. An iterative approach that uses reflection profile analysis, amplitude slice-mapping, and often both in conjunction is often necessary as neither method by itself is sufficient. In northern Australia, two constructed mounds contain a number of cultural and geological horizons and features, which can be imaged with GPR. The reflection profiles display the modified ground surface prior to mound construction and some initial construction layers. On the pre-mound surface, amplitude maps of reflective layers that were built-up on the ground surface indicate that they were constructed in an intentional manner. Those surfaces were later covered by sand to produce mounds used for human burial. Human internments in the mound can only be seen in reflection profiles, but once discovered, the profiles can be re-sliced to produce high definition amplitude images of these remains. No one method of analysis can provide an overall interpretation of these complex internal mound features. When the methods are varied, depending on the results of one method, a detailed and varied analysis of certain aspects of the mounds’ internal features are visible, leading to the generation of a number of hypotheses about how this area of northern Australia was used in the past. The 3-D data from GPR shows that this area was an important location on the landscape in the past, and was modified by the construction of a monumental mound, which was then used for human burials, and more recently, the construction of what was likely a ritual enclosure.

The research on the buried public monumental complexes of Lupiae (Lecce) by geophysical prospecting

Ongoing and extensive urbanisation may threaten important archaeological structures that are still buried in urban areas. The ground penetrating radar (GPR) method is the most promising alternative for resolving buried archaeological structures in urban territories. This paper presents a case study that involves a geophysical survey employing the surface three-dimensional (3D) GPR techniques, in order to archaeologically characterise the investigated areas. The site is located in the south-western sector of the historical centre of Lecce (Apulia, Italy), where the modern city overlaps the main public monuments of the Roman municipium of Lupiae, only partially preserved or excavated: the amphitheatre, the theatre, the baths and maybe also the Forum. GPR measurements, integrated with the results of archaeological excavations and the topographical surveys of the preserved remains, were carried out in several areas regarding sectors of the ancient roman city. The GPR data were collected along a dense network of parallel profiles. The GPR sections were processed applying specific filters to the data in order to enhance their information content. The GPR images significantly contributed in reconstructing the complex subsurface properties in these modern urban areas. Strong GPR reflections features were correlated with possible ancient structures and they were integrated in the digital archaeological map of the city.