EDGE-PRESERVING REGULARIZATION SCHEME APPLIED TO THE MODIFIED GRADIENT METHOD FOR THE RECONSTRUCTION OF TWO-DIMENSIONAL TARGETS FROM LABORATORY-CONTROLLED DATA

Abstract—In this paper, a two-dimensional inverse scattering problem dealing with microwave tomography is considered. To solve this non linear and ill-posed problem, an iterative scheme based on the Modified Gradient Method (MGM) is used. The object to be estimated is represented by a complex function, and some modifications of the MGM formulation have been considered. This algorithm leads to an efficient regularization scheme, based on edge preserving functions which act separately on the real and imaginary parts of the object. In order to show the interest of this regularized MGM, the algorithm i