GPR Full-Waveform Inversion With Deep-Learning Forward Modeling: A Case Study From Non-Destructive Testing

Ourania Patsia

doi:10.1109/TGRS.2023.3303683

2023-01-01

PRE

Ourania Patsia *

Antonios Giannopoulos

Iraklis Giannakis

DOI：10.1109/TGRS.2023.3303683

原文链接

原文求助

摘要

En 中文

Numerical modeling of ground penetrating radars (GPRs), such as the finite-difference time-domain (FDTD) method, has been extensively used to enhance the interpretation of GPR data and as a key component of full-waveform inversion (FWI). A major drawback of numerical solvers, especially within the context of FWI, is that they are still computationally expensive requiring often unattainable computational resources and access to high-performance computing (HPC). In this work, we present a near real-time deep-learning forward solver for GPR data that can generate entire B-scans, given certain model parameters as inputs. The machine-learning (ML) model is tuned for reinforced concrete slab scenarios, but the same rationale can be applied in a straightforward manner to other applications as well. The training was performed using entirely synthetic data, where a 3-D digital twin based on the 2000-MHz palm antenna from Geophysical Survey Systems, Inc. (GSSI) was included in FDTD simulations for the training set. The accuracy of the deep-learning solver is demonstrated with both synthetic and real data from reinforced concrete slabs. The predicted ML responses were in very good agreement with FDTD, showing a high degree of accuracy. The ML solver is then used as part of an FWI algorithm to characterize the concrete slab and estimate the depth and radius of the buried rebars. Coupled FWI with an ML-based forward solver results in significantly less execution times compared to conventional FWI using numerical solvers. The high accuracy of the proposed FWI, combined with the efficiency and speed of the ML-based forward solver, make the proposed scheme an ideal tool for characterizing concrete structures in nondestructive testing.

Keyword:

Concrete

deep learning

finite-difference time-domain (FDTD)

forward problem

full-waveform inversion (FWI)

ground penetrating radar (GPR)

machine learning (ML)

neural networks (NNs)