Regularization by inexact Krylov methods with applications to blind deblurring

This paper is concerned with the regularization of large-scale discrete inverse problems by means of inexact Krylov methods. Specifically, we derive two new inexact Krylov methods that can be efficiently applied to unregularized or Tikhonov-regularized least squares problems, and we study their theoretical properties, including links with their exact counterparts and strategies to monitor the amount of inexactness. We then apply the new methods to separable nonlinear inverse problems arising in blind deblurring. In this setting inexactness stems from the uncertainty in the parameters defining the blur, which may be recovered using a variable projection method leading to an inner-outer iteration scheme (i.e., one cycle of inner iterations is performed to solve one linear deblurring subproblem for any intermediate values of the blurring parameters computed by a nonlinear least squares solver). The new inexact solvers can naturally handle varying inexact blurring parameters while solving the linear deblurring subproblems, allowing for a much reduced number of total iterations and substantial computational savings with respect to their exact counterparts.

[1]  Julianne Chung,et al.  Flexible Krylov Methods for ℓp Regularization , 2018, SIAM J. Sci. Comput..

[2]  P. Hansen Discrete Inverse Problems: Insight and Algorithms , 2010 .

[3]  Paolo Novati,et al.  Inheritance of the discrete Picard condition in Krylov subspace methods , 2016 .

[4]  Valeria Simoncini,et al.  Approximating the leading singular triplets of a large matrix function , 2017 .

[5]  Zdeněk Strakoš,et al.  The regularizing effect of the Golub-Kahan iterative bidiagonalization and revealing the noise level in the data , 2009 .

[6]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[7]  J. Nagy,et al.  Numerical methods for coupled super-resolution , 2006 .

[8]  Silvia Gazzola,et al.  Krylov methods for inverse problems: Surveying classical, and introducing new, algorithmic approaches , 2020, GAMM-Mitteilungen.

[9]  Toke Koldborg Jensen,et al.  Iterative regularization with minimum-residual methods , 2007 .

[10]  James G. Nagy,et al.  Iterative Methods for Image Deblurring: A Matlab Object-Oriented Approach , 2004, Numerical Algorithms.

[11]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[12]  Yousef Saad,et al.  Iterative methods for sparse linear systems , 2003 .

[13]  Misha Elena Kilmer,et al.  A Framework for Regularization via Operator Approximation , 2015, SIAM J. Sci. Comput..

[14]  P. Alam,et al.  R , 1823, The Herodotus Encyclopedia.

[15]  James G. Nagy,et al.  Large-Scale Inverse Problems in Imaging , 2015, Handbook of Mathematical Methods in Imaging.

[16]  Valeria Simoncini,et al.  Theory of Inexact Krylov Subspace Methods and Applications to Scientific Computing , 2003, SIAM J. Sci. Comput..

[17]  Martin Hanke,et al.  On Lanczos Based Methods for the Regularization of Discrete Ill-Posed Problems , 2001 .

[18]  G. Golub,et al.  Separable nonlinear least squares: the variable projection method and its applications , 2003 .

[19]  Raymond H. Chan,et al.  A Fast Algorithm for Deblurring Models with Neumann Boundary Conditions , 1999, SIAM J. Sci. Comput..

[20]  Gerard L. G. Sleijpen,et al.  Inexact Krylov Subspace Methods for Linear Systems , 2004, SIAM J. Matrix Anal. Appl..

[21]  S. Serra-Capizzano,et al.  Improved image deblurring with anti-reflective boundary conditions and re-blurring , 2006 .

[22]  D. Calvetti,et al.  Tikhonov regularization and the L-curve for large discrete ill-posed problems , 2000 .

[23]  Jianhong Shen,et al.  Deblurring images: Matrices, spectra, and filtering , 2007, Math. Comput..

[24]  Maria Rosaria Russo,et al.  On Krylov projection methods and Tikhonov regularization , 2015 .

[25]  Per Christian Hansen,et al.  Unmatched Projector/Backprojector Pairs: Perturbation and Convergence Analysis , 2018, SIAM J. Sci. Comput..

[26]  Roland Wagner,et al.  Lanczos-based fast blind deconvolution methods , 2021, J. Comput. Appl. Math..

[27]  James G. Nagy,et al.  An Efficient Iterative Approach for Large-Scale Separable Nonlinear Inverse Problems , 2009, SIAM J. Sci. Comput..