Solução do problema inverso da tomografia por impedância elétrica utilizando o simulated annealing : uma nova abordagem

Abstract

Image reconstruction in electrical impedance tomography (EIT) deals with an ill-posed and nonlinear inverse problem. It intends to minimize the difference between simulated (virtual) object data and data from a non simulated (real) object. In this paper, a new approach to the Simulated Annealing method applied to reconstruction of EIT images is described. The main advantage in this approach is that all conductivity parameters are updated simultaneously. Other methods that employ Simulated Annealing to the problem of EIT evaluate each conductivity parameter individually resulting in high computational cost. The algorithm was tested both with computationally generated data and with measurements performed on a physical simulation tank. In both cases, the method was able to make data inversion, determining the position, the dimensions and the conductivity of materials in an opaque object plane.