Abstract:Ground penetrating radar forward simulation plays an important role in real radar data interpretation and full waveform inversion. A near real-time GPR forward simulation method based on machine learning framework is proposed for the problems of huge computation, time-consuming and unfavorable to timely detection in traditional ground-penetrating radar forward simulation. Using the detection of rebar in concrete as a GPR application scenario, the water content of concrete, the radius and the burial depth of the rebar as model parameters, and numerically simulating the scattered echo signal using FDTD. Principal component analysis is applied to reduce the dimensionality of the echo data to obtain the corresponding principal component weight coefficients, which are used as the output of the machine learning network. A multilayer recurrent network architecture and learning strategy based on random forest is designed, which not only fully explores the intrinsic causal relationship between learning model parameters and principal component weight coefficients, but also shares the interconnection between principal components, and has the function of refining and correcting for each predicted principal component, so as to realize the fast forward simulation of ground-penetrating radar based on machine learning, which effectively improves the accuracy of forward simulation compared with traditional machine learning. On this basis two deep neural networks are combined with random forest to establish a machine learning-based scene parameter prediction model with the principal component weight coefficients of echo data as input, which achieves near real-time buried target detection with a maximum error of 2% for the predicted concrete water content and 6.7% for the maximum error of the buried depth of the rebar.

Modeling of Ground Penetrating Radar System Based on Recurrent Neural Network Architecture

Study on a Recurrent Convolutional Neural Network Based FDTD Method

An Efficient Ray Tracing Based Method of Ground Penetrating Radar Simulation for Dispersive Media

Electromagnetic Modeling Using an FDTD-Equivalent Recurrent Convolution Neural Network: Accurate Computing on a Deep Learning Framework.

Finite Difference Time Domian Simulation of GPR Electromagnetic Wave

Three-dimensional numerical simulation and experimental validation of airborne ground-penetrating radar based on CNCS-FDTD method under undulating terrain conditions

Ground-penetrating radar wavefield simulation via physics-informed neural network solver

Towards 3D Metric GPR Imaging Based on DNN Noise Removal and Dielectric Estimation

Numerical simulation of GPR based on 3-D high-order FDTD

Fdtd Simulations for Ground Penetrating Radar in Urban Applications

Modeling GPR Wave Propagation in Complex Underground Structures Using Conformal ADI-FDTD Algorithm

Refined Modeling of Heterogeneous Medium for Ground-Penetrating Radar Simulation

A novel method for fast forward simulation of ground penetrating radar and buried target detection based on machine learning

2D Electromagnetic Simulation for Ground Penetrating Radar with a Topographic Ground Surface by the Curvilinear Collocated-Grid Finite-Difference Method Combined with Equivalent Field Method

GPR Simulation for the Fire Detection in Ground Coal Mine Using FDTD Method

A Novel FDTD-Based 3-D RTM Imaging Method for GPR Working on Dispersive Medium

Research on Airborne Ground-Penetrating Radar Imaging Technology in Complex Terrain

Multi-Region Finite Difference Time Domain Simulation for Airborne Radar

Numerical Simulation For Ground Penetrating Radar In Dispersive Medium

Frequency Domain 2.5d Gpr Forward Modeling

GPR-based Subsurface Object Detection and Reconstruction Using Random Motion and DepthNet