Imaging sparse metallic cylinders through a local shape function Bayesian compressive sensing approach

An innovative method for the localization of multiple sparse metallic targets is proposed. Starting from the local shape function (LSF) formulation of the inverse scattering problem and exploiting the multitask Bayesian compressive sensing (MT-BCS) paradigm, a two-step approach is described where, after a first estimation of the LSF scattering amplitudes, the reconstruction of the metallic objects is yielded through a thresholding and voting step. Selected numerical examples are presented to analyze the accuracy, the robustness, and the computational efficiency of the LSF-MT-BCS technique.