Split Bregman iterative algorithm for sparse reconstruction of electrical impedance tomography

In this paper, we present an evaluation of the use of split Bregman iterative algorithm for the L1-norm regularized inverse problem of electrical impedance tomography. Simulations are performed to validate that our algorithm is competitive in terms of the imaging quality and computational speed in comparison with several state-of-the-art algorithms. Results also indicate that in contrast to the conventional L2L2-norm regularization method and total variation (TV) regularization method, the L1-norm regularization method can sharpen the edges and is more robust against data noises.

► L1L1-norm regularization is applied for electrical impedance tomography.
► Split Bregman algorithm is used to minimize the optimal problem.
► Comparisons on different regularization models and algorithms are presented.