Complex-Valued Deep Convolutional Networks for Nonlinear Electromagnetic Inverse Scattering

Electromagnetic inverse scattering problem is a typical complex problem while traditional deep convolutional neural network can only be applied to real problem. Motivated by this, this paper presents a new approach for electromagnetic inverse problem with complex convolutional neural network. In this way, several cascaded convolutional neural network modules are introduced to learn a model to realize super-resolution for electromagnetic imaging. The simulation and experimental results show that the proposed method paves a new way addressing realtime practical large-scale electromagnetic inverse scattering problems.