Estimation of the Optimal Iteration Number for Minimal Image Discrepancy

Due to noise, the iterative image reconstruction algorithms must stop early before reaching the convergence. There is an optimal stopping point, at which the discrepancy of the reconstruction to the true image reaches minimum. It is still an open problem to find this optimal stopping point. This paper establishes two approximate relationships toward solving this open problem. The first approximate relationship is between the iterative Landweber algorithm and an iteration-number-emulated filtered backprojection (FBP) algorithm. The second approximate relationship is between the optimal iteration-number-emulated FBP reconstruction and the optimal projection-domain filtered data. These two relationships can help us to estimate the optimal stopping point.