Assessment of Material Layers in Building Walls Using GeoRadar

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