Electrical impedance tomography based on Tikhonov regularization method improved by level set method

Electrical impedance tomography (EIT) is a nonlinear and highly ill-posed inverse problem that requires using regularization technology to get solution. The generalized Tikhonov regularization method (TRM) is popular in solving the inverse problem, but its stability is sensitive to the choice of regularization parameter. It usually needs to get a balance between the accuracy and stability of the solution, so additional regularization scheme must be taken for obtaining a stable solution. In this paper, we propose a novel algorithm based on TRM which is improved by level set method (LSM). Level set here is used to represent interfaces between regions with different conductivities, and the regularization parameter can be changed adaptively during the iteration process of region boundary searching. Numerical simulation experiments have shown that the proposed algorithm obtains perfect stability and higher accuracy in reconstruction process compared with the algorithm which is based on TRM only.