Image Reconstruction Based on Sparsity and Sensitive Field Optimization for EMT

Since the sensitive fields in electromagnetic tomography (EMT) have soft-field properties, that is, the electromagnetic properties, they are easily affected by the change in medium distribution, resulting in low image reconstruction accuracy. In this article, an image reconstruction method based on sparsity and sensitive field optimization for EMT is proposed, which can improve the efficiency and accuracy. We selected appropriate sensitivity field matrix for different flow distribution and proposed a sparsity-based image reconstruction algorithm. First, a sensitivity field optimization method based on flow pattern identification is proposed. The sensitivity matrix corresponding to the flow pattern is selected according to the result of flow pattern identification. Then, based on the optimized sensitive field, a sparse image reconstruction method is proposed. For image reconstruction, the conductivity distribution vector of the EMT system is obtained by zero-padding expansion, random recombination, and orthogonal transformation, respectively. Finally, 3-D modeling and simulation experiments are carried out through COMSOL software. Simulation results have shown that this method is superior to other existing methods in imaging indexes such as image error and correlation coefficient and has better imaging effect.