Regularization of constrained maximum likelihood iterative algorithms by means of statistical stopping rule

TL;DR

This paper introduces a new statistical stopping rule for constrained maximum likelihood iterative algorithms, improving regularization in ill-posed inverse problems, especially in image reconstruction from solar flare data.

Contribution

The paper extends Tikhonov regularization to a statistical framework and demonstrates that the proposed stopping rule yields well-defined regularization methods for ISRA and EM algorithms.

Findings

The stopping rule effectively recovers input images in simulated data.

It outperforms classical stopping rules in robustness and accuracy.

Validated on real RHESSI solar flare data.

Abstract

In this paper we propose a new statistical stopping rule for constrained maximum likelihood iterative algorithms applied to ill-posed inverse problems. To this aim we extend the definition of Tikhonov regularization in a statistical framework and prove that the application of the proposed stopping rule to the Iterative Space Reconstruction Algorithm (ISRA) in the Gaussian case and Expectation Maximization (EM) in the Poisson case leads to well defined regularization methods according to the given definition. We also prove that, if an inverse problem is genuinely ill-posed in the sense of Tikhonov, the same definition is not satisfied when ISRA and EM are optimized by classical stopping rule like Morozov's discrepancy principle, Pearson's test and Poisson discrepancy principle. The stopping rule is illustrated in the case of image reconstruction from data recorded by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI). First, by using a simulated image consisting of structures analogous to those of a real solar flare we validate the fidelity and accuracy with which the proposed stopping rule recovers the input image. Second, the robustness of the method is compared with the other classical stopping rules and its advantages are shown in the case of real data recorded by RHESSI during two different flaring events.