Convolutional Neural Networks for Subsurface Electrical Properties Estimation

The estimation of electromagnetic properties of soil is a crucial step in the determination of volumetric water content (VWC). Several techniques are available for estimating VWC; however, these approaches are often time-consuming, complex, and costly. In this paper, we have developed a method for solving inverse scattering problems by utilizing a Convolutional Neural Network (CNN). This CNN takes Reflection coefficients as input and predicts soil parameters such as permittivity, conductivity, and thickness of layered soil. Our results demonstrate the effectiveness of the proposed model in accurately predicting soil parameters.