Layered medium ground penetrating radar parameter inversion based on PPAM-UNet++ network

Ground Penetrating Radar (GPR) inversion technique, as a key means to analyze radar data to accurately estimate and reconstruct the physical parameters of subsurface media, has been playing an important role in the field of geological exploration. However, most of the current studies are limited to the parameter inversion of single-layer homogeneous medium, which is difficult to fully reflect the real characteristics of the complex underground environment. In order to reflect the complex characteristics of underground medium more realistically, this paper constructs a layered medium model, adds parallelized Patch - aware attention module on the basis of UNet++ model, and proposes an advanced inversion algorithm based on PPAM-UNet++, which is fully verified in the performance of complex medium environment and proves its high effectiveness and applicability. This study provides new solution ideas for the application of ground-penetrating radar technology in more realistic underground scenes.