Back to articles
Volume: 32 | Article ID: art00002
Proximal Newton Methods for X-Ray Imaging with Non-Smooth Regularization
  DOI :  10.2352/ISSN.2470-1173.2020.14.COIMG-007  Published OnlineJanuary 2020

Non-smooth regularization is widely used in image reconstruction to eliminate the noise while preserving subtle image structures. In this work, we investigate the use of proximal Newton (PN) method to solve an optimization problem with a smooth data-fidelity term and total variation (TV) regularization arising from image reconstruction applications. Specifically, we consider a nonlinear Poisson-modeled single-energy X-ray computed tomography reconstruction problem with the data-fidelity term given by the I-divergence. The PN algorithm is compared to state-of-the-art first-order proximal algorithms, such as the wellestablished fast iterative shrinkage and thresholding algorithm (FISTA), both in terms of number of iterations and time to solutions. We discuss the key factors that influence the performance of PN, including the strength of regularization, the stopping criterion for both sub-problem and main-problem, and the use of exact or approximated Hessian operators.

Subject Areas :
Views 29
Downloads 2
 articleview.views 29
 articleview.downloads 2
  Cite this article 

Tao Ge, Umberto Villa, Ulugbek S. Kamilov, Joseph A. O’Sullivan, "Proximal Newton Methods for X-Ray Imaging with Non-Smooth Regularizationin Proc. IS&T Int’l. Symp. on Electronic Imaging: Computational Imaging XVIII,  2020,  pp 7-1 - 7-7,

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2020
Electronic Imaging
Society for Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151 USA