Titelaufnahme

Titel
An infeasible primal-dual algorithm for total bounded variation-based inf-convolution-type image restoration
Verfasser/ VerfasserinHintermüller, Michael In der Gemeinsamen Normdatei der DNB nachschlagen ; Stadler, Georg
Erschienen in
SIAM Journal on Scientific Computing, Philadelphia, Pa., 2006, Jg. 28, H. 1, S. 1-23
ErschienenSIAM
Ausgabe
Accepted version
SpracheEnglisch
DokumenttypAufsatz in einer Zeitschrift
Schlagwörter (EN)Fenchel duality / generalized Newton-type methods / image restoration / total bounded variation
ISSN1095-7197
URNurn:nbn:at:at-ubg:3-892 Persistent Identifier (URN)
DOIdoi:10.1137/040613263 
Zugriffsbeschränkung
 Das Werk ist frei verfügbar
Dateien
An infeasible primal-dual algorithm for total bounded variation-based inf-convolution-type image restoration [0.6 mb]
Links
Nachweis
Zusammenfassung (Englisch)

In this paper, a primal-dual algorithm for total bounded variation (TV)type image restoration is analyzed and tested. Analytically it turns out that employing a global Ls-regularization, with 1 < s 2, in the dual problem results in a local smoothing of the TVregularization term in the primal problem. The local smoothing can alternatively be obtained as the infimal convolution of the lr-norm, with r1 + s1 = 1, and a smooth function. In the case r = s = 2, this results in Gauss-TVtype image restoration. The globalized primal-dual algorithm introduced in this paper works with generalized derivatives, converges locally at a superlinear rate, and is stable with respect to noise in the data. In addition, it utilizes a projection technique which reduces the size of the linear system that has to be solved per iteration. A comprehensive numerical study ends the paper.

Notiz