Deep convolutional neural networks for estimating porous material parameters with ultrasound tomography

TL;DR

This paper demonstrates that deep convolutional neural networks can accurately estimate key parameters of water-saturated porous materials using ultrasound tomography data, with simulations validating the approach.

Contribution

The study introduces a novel machine learning framework combining high-order discontinuous Galerkin simulations with CNNs for porous material parameter estimation.

Findings

Neural networks accurately estimate porosity and tortuosity.

Simulation-based training confirms method's feasibility.

Less relevant parameters are effectively marginalized.

Abstract

We study the feasibility of data based machine learning applied to ultrasound tomography to estimate water-saturated porous material parameters. In this work, the data to train the neural networks is simulated by solving wave propagation in coupled poroviscoelastic-viscoelastic-acoustic media. As the forward model, we consider a high-order discontinuous Galerkin method while deep convolutional neural networks are used to solve the parameter estimation problem. In the numerical experiment, we estimate the material porosity and tortuosity while the remaining parameters which are of less interest are successfully marginalized in the neural networks-based inversion. Computational examples confirms the feasibility and accuracy of this approach.