Assesment of material layers in building walls using GeoRadar

TL;DR

This paper presents a data-driven method using GeoRadar and CNNs to non-invasively assess wall material layers by analyzing GPR data, validated on real building scans.

Contribution

It introduces a simulation-based training approach with CNNs for material layer estimation in walls, demonstrating generalization to real-world data.

Findings

CNN accurately predicts wall layer properties

Simulation data effectively trains models for real-world application

Method offers a non-invasive alternative for building assessment

Abstract

Assessing the structure of a building with non-invasive methods is an important problem. One of the possible approaches is to use GeoRadar to examine wall structures by analyzing the data obtained from the scans. We propose a data-driven approach to evaluate the material composition of a wall from its GPR radargrams. In order to generate training data, we use gprMax to model the scanning process. Using simulation data, we use a convolutional neural network to predict the thicknesses and dielectric properties of walls per layer. We evaluate the generalization abilities of the trained model on data collected from real buildings.