Inverse reconstruction algorithm using Radon space Gibbs constrain for low SNR measurements

In this paper, a new iterative method more suitable for low signal to noise (SNR) measurements was proposed. Different from conventional MAP method, we take the projection space as a Gibbs field and the penalty term we used was defined in projection space. Filtering and back-projection step were applied to generate the penalty. The spatial resolution of our method was studied and we furthermore modified our method to obtain nearly spatial invariant resolution. Future works were discussed at the end of the paper.