Enhanced Linear Sampling Method for GPR Imaging

Xiang Liu (1) , Mohammed Serhir (1) , Abelin Kameni (1) , Marc Lambert (1) , Lionel Pichon (1)

1 GeePs - Laboratoire Génie électrique et électronique de Paris

In this paper we consider the electromagnetic inverse scatting problem for GPR application. The reconstruction of the shape of buried dielectric objects is obtain using the well-known non iterative inverse technique entitled Linear Sampling Method (LSM). The applicability of LSM for GPR imaging is discussed, its performance with full and limited aperture is explored in a simple 2D free space configuration in a first step. The use of non-ideal sources and the possible influence of the antenna radiation pattern onto the results are considered. An enhanced approach is proposed to improve the robustness of the LSM with non-ideal sources.

Domaines: Electromagnétisme

Xiang Liu, Mohammed Serhir, Abelin Kameni, Marc Lambert, Lionel Pichon. Enhanced Linear Sampling Method for GPR Imaging. International Workshop on New Computational Methods for Inverse Problems (NCMIP 2016), May 2016, Cachan, France. ⟨hal-01341986⟩