An Optimized Method for Electromagnetic Inverse Scattering Imaging Based on GAN

Electromagnetic Inverse scattering problem (ISP) is aimed at recontructing shapes or retrieving electromagnetic parameters of scatterers from the scattering fields. Synthetic aperture radar methods, such as Delay-and-Sum (DAS) or Delay-Multiply-and-Sum (DMAS) are often used to solve such problems. However, the imaging results are not satisfying usually. Forthermore, these methods are associated with heavy computational cost, and consequently, they are often time-consuming. In this paper, a deep learning method named pix2pix based on conditional generative adversarial network (cGAN) is applied to enhance the quarlity of imaging and reduce the time cost. The simulation results show that the proposed method can provide a new solution for the difficult ISPs.