edoc-Server der Humboldt-Universität zu Berlin

Post- oder Preprint

Publikationsart: Artikel
Autor(en): M. Hintermüller; T. Wu
Titel: Bilevel Optimization for Calibrating Point Spread Functions in Blind Deconvolution
Erschienen in: Inverse Problems and Imaging 9 (4) 2015
S. 1139-1169
Erstveröffentlichung: 01.11.2015
Veröffentlichung auf edoc: 09.05.2017
Anmerkung: This is a pre-copy-editing, author-produced PDF of an article accepted for publication in Inverse Problems and Imaging following peer review. The definitive publisher-authenticated version (Michael Hintermüller, Tao Wu; Bilevel optimization for calibrating point spread functions in blind deconvolution; Inverse Problems and Imaging; Pages: 1139 - 1169, Volume 9, Issue 4, November 2015 doi:10.3934/ipi.2015.9.1139 ) is available online at AIMS: http://www.aimsciences.org/journals/displayArticlesnew.jsp?paperID=11746
Status: published
peer_reviewed
Volltext: pdf (urn:nbn:de:kobv:11-100246505)
URL der Erstveröffentlichung: http://www.aimsciences.org/journals/displayArticlesnew.jsp?paperID=11746
Fachgebiet(e): Mathematik
Schlagwörter (eng): blind deconvolution, bilevel optimization, MPEC, image processing
Einrichtung: Humboldt-Universität zu Berlin, Mathematisch-Naturwissenschaftliche Fakultät
Metadatenexport: Um den gesamten Metadatensatz im Endnote- oder Bibtex-Format zu speichern, klicken Sie bitte auf den entsprechenden Link. Endnote   Bibtex  
print on demand: Wenn Sie auf dieses Icon klicken, können Sie ein Druckexemplar dieser Publikation bestellen. Bestellung als gedruckte und gebundene Version bei epubli.de, Ausführung der Bestellung erst nach Bestätigung auf den epubli.de-Seiten

Abstract (eng):
Blind deconvolution problems arise in many imaging modalities, where both the underlying point spread function, which parameterizes the convolution operator, and the source image need to be identified. In this work, a novel bilevel optimization approach to blind deconvolution is proposed. The lower-level problem refers to the minimization of a total-variation model, as is typically done in non-blind image deconvolution. The upper-level objective takes into account additional statistical information depending on the partic- ular imaging modality. Bilevel problems of such type are investigated system- atically. Analytical properties of the lower-level solution mapping are established based on Robinson’s strong regularity condition. Furthermore, several stationarity conditions are derived from the variational geometry induced by the lower-level problem. Numerically, a projected-gradient-type method is employed to obtain a Clarke-type stationary point and its convergence properties are analyzed. We also implement an efficient version of the proposed algorithm and test it through the experiments on point spread function calibration and multiframe blind deconvolution.
 
Generiert am 27.05.2017, 11:13:43