Regularized reconstruction for optical tomography based on maximum entropy

A regularized reconstruction algorithm for optical tomography (OT) based on the equation of radiative transfer (ERT) is proposed with the entropy of the image as a penalty term. In this algorithm, the reconstruction problem is considered as an optimization problem that can be solved by the gradient-based image iterative reconstruction (GIIR) scheme. A gradient calculation strategy is proposed for solving the aforementioned optimization problem. Experimental results show that the algorithm proposed in this paper outperforms the traditional algorithm without such regularization.