Multi-Parameter Inversion of AIEM by Using Multi-layer and Multi-Channel Convolutional Neural Network

In this paper, a novel multi-parameter inversion method is proposed for the Advanced Integral Equation Model (AIEM) by using the multi-layer and multi-channel convolutional neural network (MMCNN). The training data consists of the full polarization scattering coefficient. The results show that the error of the inversion parameters, including the real and imaginary parts of the dielectric constant, the root mean square height and the relative length can be reduced to 2.78%, 10.13%, 1.11% and 0.55% respectively.