Inverse Problem of Multiple Objects Buried in a Half-space

Electromagnetic imaging of buried multiple dielectric cylinders by using genetic algorithm has been presented. The unknown amount and distribution of dielectric cylinders axe buried in one half-space and illuminated by transverse magnetic (TM) polarization plane wave from the other half-space. Based on the boundary condition and the measured scattered field, we have derived a set of nonlinear integral equations, and the imaging problem is reformulated into an optimization problem. In particular, by taking account into the complete nonlinear formulations, the amount and distribution of the dielectric cylinders could be highly-contrasted and complicated. In inverse algorithm, the improved steady state genetic algorithm is employed to search for the global extreme solution of objective function. Numerical results have demonstrated that the powerful performance of the inverse algorithm. Numerical results show that satisfactory reconstruction has been obtained.