Ground Penetrating Radar Technology Research Articles

A Case Study of the Integration of Ground-Based and Drone-Based Ground-Penetrating Radar (GPR) for an Archaeological Survey in Hulata (Israel): Advancements, Challenges, and Applications

This study delves into the fusion of ground-based and drone-based ground-penetrating radar (GPR) technologies in archaeological exploration. Set against the backdrop of the Hulata solar panel construction site in Israel, the research confronts daunting obstacles such as clayey soil, accurate detection of small objects, and the imperative of timely reporting crucial for construction management. The drone-based GPR, a testament to technological innovation, showcases remarkable adaptability to challenging terrains, dispelling doubts about electromagnetic wave decay in clayey soil. Methodologically, the study employs detailed orthophoto mapping and grid-type surveys. The correlation of the results significantly bolsters the reliability of archaeological discoveries, uncovering scattered artifacts buried approximately 1–1.5 m below the surface. Meticulous excavations validate the geophysical surveys, affirming the presence of structures constructed from boulders. The application at the Hulata site validates the adaptability of drone-based GPR in challenging terrains. It provides a swift, cost-effective, and minimally invasive alternative to traditional excavation techniques, thereby transforming the field of archaeology.

Field Validation of Deep-Learning-Based Ground Penetrating Radar Image Analysis for Advancing Subsurface Distress Detection

This research introduces an innovative method for detecting subsurface cracks within pavements by leveraging ground penetrating radar (GPR) technology in conjunction with advanced deep learning techniques. Its primary aim is to significantly improve the accuracy and efficiency of pavement assessment, particularly for operational and maintenance purposes. The proposed model, GPR-YOLOR (You Only Learn One Representation), extends the YOLOR framework and incorporates a region of interest within the top pavement layer to detect subsurface cracks. While the model can be trained with annotated data, the main challenge lies in validating results in the field because of the inability to visually inspect subsurface conditions and the high cost associated with direct coring. To overcome this challenge, we propose an alternative approach that utilizes the co-occurrence of surface cracks as pseudo labels, allowing for easy verification. To ensure that surface cracks correspond to subsurface cracks, the focus is exclusively on transverse cracks that develop in a bottom-up manner, such as fatigue and reflective cracks. Through this methodology, our GPR-YOLOR model achieves an F1 score of 0.72, with a precision of 0.76 and a recall of 0.68. The results from field validation underscore the effectiveness of the GPR-YOLOR model in accurately identifying subsurface cracks, highlighting its practical significance in conducting field condition assessments.

Underground utility inspection using ground penetrating radar

Underground utility refers to any below-ground line, installation, or structure used by a service or utility provider. Underground utility inspection is a process of examining the designated area that needs to be inspected in detail. The purpose of the inspection is to ensure that the underground utility is in good condition and to check for any faults. By using the ground penetrating radar (GPR), the technology can capture images below the surface of the ground using radio waves. The various data through previous references are used to discuss and analyse the capability of GPR for underground utility inspection to ensure its gives the best performance for the inspection. GPR technology has become an essential tool for locating buried utilities and preventing damage and accidents during construction projects. This paper explores various advancements in GPR technology, including data processing methods and scanning techniques, to improve the accuracy and effectiveness of underground utility mapping. The research presented in this paper highlights the abilities of GPR to inspect underground utilities. The GPR data used in this project was collected from the evaluation and verification facility (EVF) in the Malaysia Nuclear Agency. The accuracy of these predicted positions was validated from GPR data and compared with as-built drawings from the contractor. The alternative analysis approaches that could be applied to improve the GPR accuracy when the optimum conditions are not met are also discussed.

Development of Aggregate Dielectric Constant Database Protocol for Asphalt Concrete Density Prediction

Contractors are paid incentives/disincentives based on achieved in situ asphalt concrete (AC) density. Ground-penetrating radar (GPR) has been proven feasible for predicting in situ density of AC pavements using various empirical and fundamental approaches. However, to use fundamental equations for density prediction, aggregate dielectric constant must be known beforehand. Destructive cores are usually extracted from the pavement to back-calculate the aggregate dielectric constant. This cancels out the non-destructive benefit of the GPR technology. In this study, GPR is used to quantify the dielectric constant of 10 different aggregates commonly used in AC mixes in Illinois, U.S. The sampled aggregates included limestone, dolomite, trap rock, granite, and crushed gravel. The purpose was to initiate an aggregate dielectric constant database that would help predict AC density nondestructively and accurately. By using the aggregate database, GPR would help contractors estimate achieved density in real-time without prior calibration. This technology would save energy, time, and cost. Simulations using gprMax and a sensitivity analysis are presented to illustrate the effect of aggregate dielectric constant on AC bulk dielectric constant and, consequently, on AC density predicted by the Al-Qadi Lahouar Leng model. A new procedure to determine the aggregate dielectric constant using the electromagnetic mixing theory is detailed. Advanced chemical tests confirmed that the aggregate dielectric constant is a function of its elemental/mineral compositions. Finally, laboratory and field data were used to validate AC density prediction using the aggregate database. The establishment of an aggregate dielectric constant database would improve the accuracy of GPR in nondestructive AC density prediction.

CycleGAN-Based Data Augmentation for Subgrade Disease Detection in GPR Images with YOLOv5

Vehicle-mounted ground-penetrating radar (GPR) technology is an effective means of detecting railway subgrade diseases. However, existing methods of GPR data interpretation largely rely on manual identification, which is not only inefficient but also highly subjective. This paper proposes a semi-supervised deep learning method to identify railway subgrade diseases. This method addresses the sample imbalance problem in the defect dataset by utilizing a data augmentation method based on a generative adversarial network model. An initial network model for disease identification is obtained by training the YOLOv5 network with a small number of existing samples. The intelligently extended samples are then labeled to achieve a balance in the disease samples. The network is trained to improve the recognition accuracy of the intelligent model using a more complete dataset. The experimental results show that the accuracy of the proposed method can reach up to 94.53%, which is 23.85% higher than that of the supervised learning model without an extended dataset. This has strong industrial application value for railway subgrade disease detection as the potential learning ability of the model can be explored to a greater extent, thereby improving the recognition accuracy of subgrade diseases.

Assessment of In Situ Compactness and Air Void Content of New Asphalt Layers Using Ground-Penetrating Radar Measurements

This paper deals with the possibilities of ground-penetrating radar (GPR)-based quality control testing, which was demonstrated on an experimental road section of a ~220–240 m long Hungarian residential street. The measurements and their assessment aimed to control the prescribed compactness and air void content of newly built asphalt layers. Research has discussed the relationship between the air void content and the dielectric constant of asphalt layers, and provided empirical results for this relationship. We suggest a new logistic model with lower and upper asymptotes instead of the exponential formula often used in the literature. Contrary to this newly developed robust model, existing models are sensitive to extreme dielectric constant values due to the mathematical nature of their exponential function. The results of the new logistic model are compared to those of the Hoegh–Dai (HD) and Minnesota Department of Transportation (MnDOT) models on the basis of a few calibration data points. Through systematic data collection and analysis, the developed robust empirical model demonstrates a significant correlation between the relative permittivity and air void content in asphalt mixes, enabling accurate estimation of the air void content within a ±0.5% margin of error. The air void content can be applied to estimate the asphalt layer modulus. The developed model can be further exploited by utilizing a combination of GPR and drone technology. The “symbiosis” of these technologies can lead to a totally non-destructive imaging system, which can then be applied to environmental monitoring of roads and their surroundings in terms of quality control of asphalt compaction work and the hot asphalt mix behind the compaction roller during pavement construction.

Research on the Technique of Determining the Internal Structure of Ancient Stone Pagodas Based on the Discriminant Features of Ground Penetrating Radar

Abstract Obtaining detailed information about the design and construction of ancient stone pagodas is challenging due to their significant age. Due to this, the internal structure is still unknown, and the analysis of their structural safety lacks a solid foundation. The purpose of this paper is to research ancient stone pagodas and propose adequate methods for identifying their internal structure and assessing their damage status using ground-penetrating radar nondestructive detection technology. The internal structure of the pagoda is inferred through a comprehensive process of layer-by-layer scanning using shielded antennas of varying frequencies, incorporating distance correction. The results demonstrate that by selecting suitable radar antenna frequencies, both test profiles become visible, and the position and characteristics of the internal substantial reflection area exhibit a high degree of consistency, mutually reinforcing each other. The White Pagoda in Hangzhou is believed to have both a mortise and tenon joint structure and a column structure. The ground penetrating radar detection method proves to be effective and has valuable applications.

Exploring spatial distribution characteristics of Populus simonii Carr coarse roots using ground penetrating radar

A comprehensive root database holds paramount importance for agro-forestry endeavours, providing valuable insights into the ecological characteristics and stability of plant species. However, the creation of such databases presents a formidable challenge. This study is devoted to shedding light on the spatial distribution characteristics of coarse roots (diameter ≥1 cm) in Populus simonii trees, which naturally flourish in Tongde County, Thanh Hai province, China. Ground-penetrating radar (GPR) technology was employed to investigate the distribution of these coarse roots, while tree root diameters were estimated through controlled experiments involving buried objects. A total of 40 P. simonii trees were meticulously selected for data collection, and the coarse roots of these trees were categorized into three diameter classes: 1-2 cm, 2-3 cm, and ≥3 cm. The study findings have unveiled a robust correlation between the density of coarse roots and the diameter at breast height (dbh) of the trees. Moreover, the total coarse root density moderately correlates with canopy width and tree height within a horizontal plane extending 3-5 m from the tree base and a vertical plane at a soil depth of 20-40 cm. This study underscores the effectiveness of GPR in elucidating the spatial distribution of tree roots and accurately estimating root diameters. This underscores the potential of GPR to significantly enhance investigations related to root ecology and below-ground biomass estimation.

A Refined Combined Grid Model for Characterizing Concealed Microcracks with Various Geometric Shapes Based on Radar Signal Processing

The concealed microcracks in shield tunnel lining present the characteristics of being of small size, unknown shape, and are difficult to detect. Based on the finite-difference time domain (FDTD) approach, this study proposed a new construction method of a refined grid accommodating and combining the variable shapes of microcracks, and capable of designing cross type, mesh type, and wave type microcrack models. The proposed new method also configured steel bars in the models to simulate actual engineering conditions, and characteristic response images of the models under different working conditions were obtained using ground penetrating radar (GPR) technology, which were then compared and analyzed to identify the imaging characteristics and differences of microcracks with variable geometric shapes. The waveform, amplitude, and time span of the characteristic single channel signal were furthermore studied. The results showed that the new method could successfully simulate the GPR characteristic response images of 0.5[Formula: see text]mm microcracks of diverse geometric shapes. When the microcracks were wavy, their real shape could only be determined after signal pre-processing; the density and quantity of steel bars directly affected the appearance of microcrack characteristic signals; the greater the density and quantity of steel bars, the greater the interference on the waveform, amplitude, and time-frequency range of electromagnetic wave signals; a special correlation existed between the maximum mean root square value of the amplitude and the single channel signal of the cracks. Moreover, the finding that the extension in time and distance in the GPR time distance profile intersected with the cracks was deemed potentially to provide fresh insights into identifying the characteristic points of the cracks in the GPR images. The new method proposed in this study successfully obtained the GPR numerical simulation images and characteristic signals of microcracks with variable geometric shapes. Through the processing and analysis of the characteristic response signals of microcracks, the conclusions obtained were considered to provide an interpretation basis for the detection of microcracks in practical engineering.

Using Ground Penetrating Radar (GPR) to Predict Log Moisture Content of Commercially Important Canadian Softwoods

The non-destructive testing of wood fibre properties is crucial for informing forest management decisions and achieving optimal resource utilization. Moisture content (MC) is an important indicator of wood freshness and may reveal the presence of wood degradation. However, efficient methods are still needed to better monitor this property along the forest–wood value chain. The objective of the study was to develop prediction models to evaluate log MC based on the propagation of ground penetrating radar (GPR) signals. A total of 165 trees representing four species (black spruce (Picea mariana (Mill.) B.S.P.), white spruce (Picea glauca (Moench) Voss), red spruce (Picea rubens Sarg.), and balsam fir (Abies balsamea (L.) Mill.)) were harvested in two regions of the province of Quebec. GPR signals were acquired in the green (fresh) state and at three subsequent drying stages. Partial least squares regression (PLSR) and locally weighted PLSR (LWPLSR) were employed to establish relationships between GPR signals (antenna frequency: 1.6 GHz) and log properties. The models were fitted on three calibration sets containing four drying stages and different species mixes. The LWPLSR models performed better than the PLSR models for predicting log MC, with a lower root mean square error (RMSEp range: 10.8%–20.2% vs. 13.0%–20.5%) and a higher R2p (0.63–0.87 vs. 0.62–0.82). Spruce-only models performed considerably better than fir-only models while multi-species models were in-between. Despite the complex anisotropy of wood and the physics of wave propagation, the GPR technology can be successfully used to estimate log moisture content, but the GPR-based MC models should be calibrated for each specific type of wood material.

A Case Study on Detection of Cavity in Urban Roads Using 3D Ground Penetrating Radar

In recent years, urban roads collapse accidents often occur, which seriously threatens the safety of people’s lives and property and the order of urban operation, and has a huge social impact. Therefore, it is very necessary to carry out road cavity detection and eliminate the hidden dangers of road operation safety. Three dimensional ground penetrating radar (GPR) technology is a new technology developed in recent years. At present, it has been widely used in road detection. This paper briefly introduces the process of urban roads detection using 3D GPR, and introduces an example of cavity detection. The results show that the effect of 3D GPR in road cavity detection is remarkable, which is worth popularizing.

Modeling and Implementation of a Joint Airborne Ground Penetrating Radar and Magnetometer System for Landmine Detection

We modeled and implemented a joint airborne system integrating ground penetrating radar (GPR) and magnetometer (MAG) models specifically for landmine detection applications. We conducted both simulations and experimental analyses of the joint airborne GPR and MAG models, with a focus on detecting the metallic components of different types of landmines, including antitank (AT) M15 metallic, antipersonnel (AP) M16 metallic, and AT M19 plastic (minimum-metal) landmines. The GPR model employed the finite-difference time-domain (FDTD) method and was evaluated using a singular value decomposition (SVD) and Kirchhoff migration (KM) with matched filtering (MF). These advanced techniques enabled the automatic identification and precise focusing of the reflected hyperbolic signals emitted by the landmines while considering cross-range resolution. Additionally, the MAGs were utilized based on the magnetic dipole model with a de-trend and a spatial median filtering method to estimate the magnetic anomaly of the landmines while considering various data spatial intervals. The joint airborne GPR and MAG system was implemented by combining and integrating the GPR and MAG models for experimental validation. Through this comprehensive approach, which included experiments, simulations, and data processing, the design parameters of the final system were obtained. These design parameters can be used in the development and application of landmine detection systems based on airborne GPR and MAG technology.

Morphology Dynamics of Ice Cover in a River Bend Revealed by the UAV-GPR and Sentinel-2

After the formation of the bend ice cover, the ice thickness of the bend is not uniformly distributed, and an open-water area is usually formed downstream of the bend. The spatial and temporal variation of the ice thickness in seven cross sections was determined via Unmanned Aerial Vehicle Ground Penetrating Radar (UAV-GPR) technology and traditional borehole measurements. The plane morphology change of the open water was observed by Sentinel-2. The results show that the average dielectric permittivity of GPR was 3.231, 3.249, and 3.317 on three surveys (5 January 2022, 16 February 2022, and 25 February 2022) of the Yellow River ice growing period, respectively. The average ice thickness of the three surveys was 0.402 m, 0.509 m, and 0.633 m, respectively. The ice thickness of the concave bank was larger than that of the convex bank. The plane morphology of the open water first shrinks rapidly longitudinally and then shrinks slowly transversely. The vertical boundary of the open water was composed of two arcs, in which the slope of Arc I (close to the water surface) was steeper than that of Arc II, and the hazardous distance of the open-water boundary was 10.3 m. The increased flow mostly affected the slope change of Arc I. Finally, we discuss the variation of hummocky ice and flat ice in GPR images and the physical factors affecting GPR detection accuracy, as well as the ice-thickness variation of concave and convex banks in relation to channel curvature.

Practice of geo-radar research in the Republic of Kazakhstan

The article considers the issues of applying ground penetrating radar (GPR) technologies for engineering purposes concerning subsurface research. It describes the field of application of GPR research during road construction and operation and the advantages and disadvantages of this type of engineering survey. It presents comparative data on the accuracy (inaccuracy) of existing GPR control methods applied to determine the thickness of the monolithic pavement layers in Europe, CIS, and the USA. The main provisions of the GPR survey procedure are described, including four main stages: the analysis of initial materials of surveyed section and equipment preparation; GPR survey; geological verification; processing and interpreting of radargrams; and the preparation of a report. Geophysical works were performed using the geo-radar of the OKO series as part of the road measuring complex DVK-05 on the section of the Astana-Petropavlovsk A-1 highway with cement concrete pavement and on the section of the R-12 "Kokshetau - Atbasar" with asphalt pavement. The example of a radargram and the core sample of a cement concrete pavement taken during geological verification of the thickness of a monolithic layer is presented. Graphs of variation in the thickness of pavement layers by radargrams of longitudinal passages in the indicated road sections with the assumptions about the nature of the heterogeneity of the obtained values are given. The obtained results on the thickness of monolithic layers (cement concrete and asphalt concrete) were assessed on the criteria of quantitative deviation from the required standard value. Recommendations were provided to address positive deviations in the thickness of monolithic pavement layers at various stages of the road survey.

USE OF GPR TECHNOLOGIES IN THE ROAD PAVEMENT SURVEY

The article contains information on the study of the possibilities of introducing GPR technologies into the practice of road surveying. The prospects for the development of this scientific direction are also analyzed. The tasks that can be solved with the use of ground-penetrating radar equipment during the non - rigid pavement. The main reasons preventing a more dynamic implementation of ground-penetrating radar technologies in the construction industry are considered in detail. The main reasons are: the difficulty of interpreting GPR data, the lack of an instrument base, the lack of a single complex of equipment and mathematical support for non - rigid pavement, which allows for guaranteed high-precision interpretation of the results of GPR surveys of non - rigid pavement and ground surfaces, and the absence of a regulatory framework that regulates the survey of highways with the use of ground-penetrating radars. Modeling of signal propagation in an environment with known electrophysical characteristics allows to determine the shape and amplitude of a complete signal, giving the possibility of correct interpretation of radargrams in order to determine the speed of signal propagation in the studied structures, electrophysical characteristics of structural materials of layers of non - rigid pavement, the interface of environments. To carry out experimental studies aimed at the development of methods of thickness measurement and defectoscopy of structural layers of non - rigid pavement, specialists of the Kharkiv national automobile and highway university use a set of geolocation equipment that uses ultra-broadband signals. A set of georadar equipment is used both for conducting experimental laboratory research of road clothing and for field surveys for the purpose of developing a road monitoring system using radar methods. For the verification and practical use of the proposed scheme for determining the time delay of signals, which allows to calculate the thickness of the structure under study, appropriate software was developed. A series of laboratory experiments were conducted using this software product. As shown by the experimental laboratory studies, the method allows to restore the thickness of the layer under investigation with a fairly high accuracy. Keywords: GPR diagnostics, highway, non - rigid pavement, ground surface, reliability, operational condition, research, electrophysical properties.

High-Throughput Canopy and Belowground Phenotyping of a Set of Peanut CSSLs Detects Lines with Increased Pod Weight and Foliar Disease Tolerance

We deployed field-based high-throughput phenotyping (HTP) techniques to acquire trait data for a subset of a peanut chromosome segment substitution line (CSSL) population. Sensors mounted on an unmanned aerial vehicle (UAV) were used to derive various vegetative indices as well as canopy temperatures. A combination of aerial imaging and manual scoring showed that CSSL 100, CSSL 84, CSSL 111, and CSSL 15 had remarkably low tomato spotted wilt virus (TSWV) incidence, a devastating disease in South Georgia, USA. The four lines also performed well under leaf spot pressure. The vegetative indices showed strong correlations of up to 0.94 with visual disease scores, indicating that aerial phenotyping is a reliable way of selecting under disease pressure. Since the yield components of peanut are below the soil surface, we deployed ground penetrating radar (GPR) technology to detect pods non-destructively. Moderate correlations of up to 0.5 between pod weight and data acquired from GPR signals were observed. Both the manually acquired pod data and GPR variables highlighted the three lines, CSSL 84, CSSL 100, and CSSL 111, as the best-performing lines, with pod weights comparable to the cultivated check Tifguard. Through the combined application of manual and HTP techniques, this study reinforces the premise that chromosome segments from peanut wild relatives may be a potential source of valuable agronomic traits.

Variability in mollic epipedon thickness in response to soil erosion–deposition rates along slopes in Northeast China

The Mollisol region of Northeast China is one of the most significant grain-producing areas. Intense soil erosion has led to a sharp decline in both the mollic epipedon (ME) thickness and soil quality, and these processes have in turn resulted in a serious threat to the national food security. Soil erosion–deposition processes influence the soil thickness. However, the quantitative relationship between the ME thickness and soil erosion–deposition rate along slopes remains unclear. In this study, the ME thickness and soil erosion–deposition rate along three typical long, gentle slopes in Heilongjiang Province were investigated by using ground-penetrating radar (GPR) and 137Cs tracing technology. Variations in the spatial characteristics of the ME thickness and soil erosion–deposition rate on slopes were obtained. Wavelet analysis was employed to analyse the soil erosion–deposition rate on these slopes. The results indicated that the soil erosion rate along the three slopes was periodic and oscillated between high and low values. The range of primary cycle variation was 73–80 m, and the range of secondary cycle variation was 44–47 m. A negative correlation was found between the soil erosion–deposition rate and ME thickness within each cycle. The periodic variability in the soil erosion rate on these slopes also yielded a periodic variability in the ME thickness. Fitted equations were derived based on the ME thickness, slope length and soil erosion–deposition rate. The error range for soil thickness prediction was 0.30–9.40 %. This study could provide an efficient and accurate method to predict ME thickness change patterns along slopes and to help arrange soil erosion control measures at appropriate locations.

Ground-Penetrating Radar Method for Studying Water Drainage in Sand Layers

A ground-penetrating radar (GPR) technology was developed to study the process of water drainage in sand layers with an insignificant concentration of dusty and clayey particles when moistened from above. The technology includes a method of calibration of the GPR equipment, algorithms for processing the GPR information, and their software implementation. The technology was used to process the results of laboratory GPR measurements obtained during draining of water through sand layers from different quarries for 100 h. The absolute values and the changes in the refractive index and specific conductivity near the sand layer upper boundary and on average over the layer depth were calculated. The results show that the developed technology makes it possible to determine electrophysical properties with an accuracy of up to 10%. The developed method for calculating relative reflectivity and its derivative with respect to the depth of the layer made it possible to visualize the information contained in the radargrams on the distribution of water near the surface and deep in the sand layers. The application of the method makes it possible to quantitatively estimate the moisture content near the upper boundary of the layer and the depth of the location of the most moistened areas of the layer depending on the duration of water drainage.

Research on Real-Time Automatic Picking of Ground-Penetrating Radar Image Features by Using Machine Learning

Hard foreign objects such as bricks, wood, metal materials, and plastics in orchard soil can affect the operational safety of garden machinery. Ground-Penetrating Radar (GPR) is widely used for the detection of hard foreign objects in soil due to its advantages of non-destructive detection (NDT), easy portability, and high efficiency. At present, the degree of automatic identification applied in soil-oriented foreign object detection based on GPR falls short of the industry’s expectations. To further enhance the accuracy and efficiency of soil-oriented foreign object detection, we combined GPR and intelligent technology to conduct research on three aspects: acquiring real-time GPR images, using the YOLOv5 algorithm for real-time target detection and the coordinate positioning of GPR images, and the construction of a detection system based on ground-penetrating radar and the YOLOv5 algorithm that automatically detects target characteristic curves in ground-penetrating radar images. In addition, taking five groups of test results of detecting different diameters of rebar inside the soil as an example, the obtained average error of detecting the depth of rebar using the detection system is within 0.02 m, and the error of detecting rebar along the measuring line direction from the location of the starting point of GPR detection is within 0.08 m. The experimental results show that the detection system is important for identifying and positioning foreign objects inside the soil.

Study of in situ dynamic modulus prediction of asphalt mixture utilizing Ground penetrating radar technology

• The small-scale specimen’s applicability in dynamic modulus test was investigated. • Empirical and ANN dynamic modulus prediction models were proposed using GPR data. • The proposed models provided acceptable results at the confidence level of 95%. It is widely known that the asphalt mixture dynamic modulus is the key mechanical property that affects the quality and service life of asphalt pavement greatly. Ground penetrating radar (GPR) technology is a prospective method commonly used in determining the material information of asphalt pavement. Its utilization is supported by studying the relationship between the characterizations of GPR waveform and the parameters representing the condition of asphalt pavement. However, it is rarely applied in dynamic modulus prediction of asphalt mixture. To fill this gap, this study focuses on developing the corresponding dynamic modulus predictive method on the basis of GPR data. First, the applicability of the ϕ 38 mm × h 110 mm small-scale specimen covering wide range of air void content in the uniaxial compression test was analyzed. Then, the empirical and artificial neural network (ANN) dynamic modulus prediction models were developed by taking into account the dielectric constant of asphalt mixture. Last, the two proposed models were validated based on four asphalt mixture lanes in the field. It was found that the test errors of small-scale specimens were smaller than 15 %, and the statistical difference of dynamic moduli and phase angles measured from the full-size and small-scale specimens is insignificant at the 95 % confidence level meaning the size influence can be ignored. According to the verification experiment, the proposed empirical model and ANN model could yield acceptable results with the average prediction error of 19.3 % and 24.8 %, respectively. It is recommended that the two proposed predictive models coupled with GPR data could be used as an alternative to the currently available methods in estimating the dynamic modulus of asphalt mixture in situ.