A New Steplength Selection for Scaled Gradient Methods with Application to Image Deblurring
- 256 Downloads
Gradient methods are frequently used in large scale image deblurring problems since they avoid the onerous computation of the Hessian matrix of the objective function. Second order information is typically sought by a clever choice of the steplength parameter defining the descent direction, as in the case of the well-known Barzilai and Borwein rules. In a recent paper, a strategy for the steplength selection approximating the inverse of some eigenvalues of the Hessian matrix has been proposed for gradient methods applied to unconstrained minimization problems. In the quadratic case, this approach is based on a Lanczos process applied every \(m\) iterations to the matrix of the gradients computed in the previous \(m\) iterations, but the idea can be extended to a general objective function. In this paper we extend this rule to the case of scaled gradient projection methods applied to constrained minimization problems, and we test the effectiveness of the proposed strategy in image deblurring problems in both the presence and the absence of an explicit edge-preserving regularization term.
KeywordsImage deconvolution Constrained optimization Scaled gradient projection methods Ritz values
Mathematics Subject Classification65K05 65R32 68U10 90C06
This work has been partially supported by the Italian Spinner 2013 Ph.D. Project “High-complexity inverse problems in biomedical applications and social systems” and by MIUR (Italian Ministry for University and Research), under the projects FIRB—Futuro in Ricerca 2012, contract RBFR12M3AC, and PRIN 2012, contract 2012MTE38N. The Italian GNCS—INdAM (Gruppo Nazionale per il Calcolo Scientifico—Istituto Nazionale di Alta Matematica) is also acknowledged.
- 5.Bertero, M., Lantéri, H., Zanni, L.: Iterative image reconstruction: a point of view. In: Censor, Y., Jiang, M., Louis, A.K. (eds.) Mathematical Methods in Biomedical Imaging and Intensity-Modulated Radiation Therapy, pp. 37–63. Edizioni della Normale, Pisa (2008)Google Scholar
- 7.Bertsekas, D.: Convex Optimization Theory. Supplementary Chapter 6 on Convex Optimization Algorithms. Athena Scientific, Belmont (2009)Google Scholar
- 12.Bonettini, S., Prato, M.: A new general framework for gradient projection methods (2014). arXiv:1406.6601
- 16.Carlavan, M., Blanc-Féraud, L.: Regularizing parameter estimation for Poisson noisy image restoration. In: International ICST Workshop on New Computational Methods for Inverse Problems, May 2011, Paris, FranceGoogle Scholar