EMT Image Reconstruction With Observation Equation Recombination and PSF Fitting

Electromagnetic tomography (EMT) is commonly applied for monitoring the distribution of conductive or magnetic materials. However, recent technologies are limited by the lack of prior information and the blurring problem of reconstructed images. In this article, an EMT image reconstruction scheme with observation equation recombination and point spread function (PSF) fitting is proposed. First, to realize conductivity distribution estimation efficiently with insufficient prior information, the observation equation is recombined by zero-padding and random arrangement with the extracted frequency information. The conductive distribution can be reconstructed by solving an L0 norm optimization problem. Then, to achieve edge deblurring of the reconstructed image, a Lorenz deblurring model for PSF fitting is proposed. Combining with Lucy-Richardson (L-R) algorithm, the EMT image is reconstructed and optimized. Finally, simulation experiments have shown the superiority of the proposed method on correlation coefficients and relative errors. It can effectively suppress the reconstruction effect caused by edge blur and has anti-noise performance.