A Gans-Based Deep Learning Framework for Automatic Subsurface Object Recognition From Ground Penetrating Radar Data

Xin, Zhang; Han, Liangxiu; Robinson, M. S.; Gallagher, Anthony

doi:10.1109/access.2021.3064205

Cited by 36 publications

(18 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data samples generated can effectively improve the identification of subsurface defects and objects. 120 However, this type of research is still relatively rare. Further development of the GAN structure and training strategies is required to provide more realistic scanned images.…”

Section: Ground Penetrating Radar-oriented Methodsmentioning

confidence: 99%

Review of artificial intelligence-based bridge damage detection

Zhang

Yuen

2022

Advances in Mechanical Engineering

View full text Add to dashboard Cite

Bridges are often located in harsh environments and are thus extremely susceptible to damage. If the initial damage is not detected in time, it can develop further causing safety hazards. Therefore, accurate detection of bridge damage is an important topic. In recent years, artificial intelligence technology has been developed rapidly, especially machine learning algorithms, which have shown amazing results in various fields while it also received attention in bridge inspection. This paper summarizes the progress of bridge damage detection research related to artificial intelligence techniques between 2015 and 2021. For structural health monitoring, sensing data is the basis for various data processing methods. The strength and weakness of the sensing data itself directly affect the effectiveness of subsequent processing methods. As a result, this paper classifies bridge damage detection studies into six categories from the types of sensing data: visual image, point cloud, infrared thermal imaging, ground-penetrating radar, vibration response, and other types of data. These six types of damage detection methods were reviewed and summarized respectively. Finally, challenges and future trends were discussed.

show abstract

Section: Ground Penetrating Radar-oriented Methodsmentioning

confidence: 99%

Review of artificial intelligence-based bridge damage detection

Zhang

Yuen

2022

Advances in Mechanical Engineering

View full text Add to dashboard Cite

show abstract

“…Similar to seismic in oil and gas exploration, interpretation of GPR data is based primarily on reflector patterns, using radar facies to analyze rock formations in the underground medium [19,24,25]. The interface and internal structure of a geological body are interpreted in terms of the amplitude, configuration, and continuity of the reflector and its external geometry [26][27][28].…”

Section: Interpretation Of Gpr Datamentioning

confidence: 99%

3D Sedimentary Architecture of Sandy Braided River, Based on Outcrop, Unmanned Aerial Vehicle and Ground Penetrating Radar Data

et al. 2022

View full text Add to dashboard Cite

Ground Penetrating Radar (GPR) is a geophysical method that uses antennas to transmit and receive high-frequency electromagnetic waves to detect the properties and distribution of materials in media. In this paper, geological observation, UAV detection and GPR technology are combined to study the recent sediments of the Yungang braided river study area in Datong. The application of the GPR technique to the description of fluvial facies and reservoir architecture and the development of geological models are discussed. The process of GPR detection technology and application includes three parts: GPR data acquisition, data processing and integrated interpretation of GPR data. The geological surface at different depths and scales can be identified by using different combinations of frequencies and antenna configurations during acquisition. Based on outcrop observation and lithofacies analysis, the Yandong Member of the Middle Jurassic Yungang Formation in the Datong Basin has been identified as a typical sandy braided river sedimentary system. The sandy braided river sandbody changes rapidly laterally, and the spatial distribution and internal structure of the reservoir are very complex, which has a very important impact on the migration and distribution of oil and gas as a reservoir. It is very important to make clear the characteristics of each architectural unit of the fluvial sand body and quantitatively characterize them. The architectural elements of the braided river sedimentary reservoir in the Datong-Yungang area can be divided into three types: channel unit, bar unit and overbank assemblages. The geological radar response characteristics of different types of sedimentary units are summarized and their interfaces are identified. The channel sediments form a lens-shaped wave reflection with a flat at the top and convex-down at the bottom in the radar profile, and the angles of the radar reflection directional axes are different on both sides of the sedimentary interface. In the radar profile, the deposit of the unit bar is an upward convex reflection structure. The overbank siltation shows a weak amplitude parallel reflection structure. The flood plain sediments are distributed continuously and stably in the radar profile, showing weak reflection characteristics. Different sedimentary units are identified by GPR data and combined with Unmanned Aerial Vehicle (UAV) detection data, and the establishment of the field outcrop geological model is completed. The development pattern of the diara is clarified, and the swing and migration of the channel in different stages are identified.

show abstract

“…The response features extracted from the peaks and their locations within the subsequent 1D time-varying amplitude signals by means of principle component analysis (PCA), are employed to classify the material type into three groups by using k-nearest neighbor supervised learning classification algorithm 26 . Other Artificial Intelligence (AI) algorithms have been successfully used for buried target recognition in GPR images include Deep Learning (DL), especially Convolutional Neural Network (CNN) frameworks 14 – 16 , 18 , 19 , 21 . 3D GPR data generated along longitudinal and cross axes is analyzed in CNN and LSTM (Long Short-Term Memory) units combined into a framework of a cascaded structure for the detection of buried explosive objects and discrimination targets or non-target alarms 15 .…”

Section: Introductionmentioning

confidence: 99%

Buried object characterization by data-driven surrogates and regression-enabled hyperbolic signature extraction

Yurt

Torpi

Kızılay

et al. 2023

Sci Rep

View full text Add to dashboard Cite

This work addresses artificial-intelligence-based buried object characterization using FDTD-based electromagnetic simulation toolbox of a Ground Penetrating Radar (GPR) to generate B-scan data. In data collection, FDTD-based simulation tool, gprMax is used. The task is to estimate geophysical parameters of a cylindrical shape object of various radii, buried at different positions in the dry soil medium simultaneously and independently of each other. The proposed methodology capitalizes on a fast and accurate data-driven surrogate model developed for object characterization in terms of its vertical and lateral position, and the size. The surrogate is constructed in a computationally efficient manner as compared to methodologies using 2D B-scan image. This is achieved by operating at the level of hyperbolic signatures extracted from the B-scan data through linear regression, which effectively reduces the dimensionality and the size of data. The proposed methodology relies on reducing of 2D B-scan image to 1D data including variation of reflected electric fields’ amplitudes with respect to the scanning aperture. The input of the surrogate model is the extracted hyperbolic signature obtained through linear regression executed on the background subtracted B-scan profiles. The hyperbolic signatures encode information about the geophysical parameters of the buried object, including depth, lateral position, and radius, all of which can be extracted using proposed methodology. Parametric estimation of the object radius and the estimation of the location parameters simultaneously is a challenging problem. Applying the application of processing steps on B-scan profiles incurs high computational costs, which is a limitation of the current methodologies. The metamodel itself is rendered using a novel deep-learning-based modified multilayer perceptron (M2LP) framework. The presented object characterization technique is favourably benchmarked against the state-of-the-art regression techniques, including Multilayer Perceptron (MLP), Support Vector Regression Machine (SVRM), and Convolutional Neural Network (CNN). The verification results demonstrate the average mean absolute error of 10 mm, and the average relative error of 8 percent, both corroborating the relevance of the proposed M2LP framework. In addition, the presented methodology provides a well-structured relation between the geophysical parameters of object and the extracted hyperbolic signatures. For the sake of supplementary verification under realistic scenarios, it is also applied for scenarios involving noisy data. The environmental and internal noise of the GPR system and their effect is analyzed as well. Furthermore, the proposed surrogate modeling approach is validated using measurement data, which is indicative of suitability of the approach to handle physical measurements as data sources.

show abstract

A Gans-Based Deep Learning Framework for Automatic Subsurface Object Recognition From Ground Penetrating Radar Data

Cited by 36 publications

References 29 publications

Review of artificial intelligence-based bridge damage detection

Review of artificial intelligence-based bridge damage detection

3D Sedimentary Architecture of Sandy Braided River, Based on Outcrop, Unmanned Aerial Vehicle and Ground Penetrating Radar Data

Buried object characterization by data-driven surrogates and regression-enabled hyperbolic signature extraction

Contact Info

Product

Resources

About