A Linear Sampling Imaging Scheme based on Convolutional Neural Network

A novel inverse scattering imaging scheme based on deep learning is proposed. Based on Linear Sampling Method (LSM), a reasonable threshold selection strategy is set to solve the contrast distribution of the target scatterer, which is taken as the input of network. The convolutional neural network is used to simulate the electromagnetic scattering process and extract the high-dimensional fine features of contrast distribution, so as to achieve accurate imaging of target scatterers.