An Adaptive Correlated Image Prior for Image Restoration Problems

0490175 - ÚTIA 2019 RIV US eng J - Článek v odborném periodiku
Ševčík, J. - Šmídl, Václav - Šroubek, Filip
An Adaptive Correlated Image Prior for Image Restoration Problems.
IEEE Signal Processing Letters. Roč. 25, č. 7 (2018), s. 1024-1028. ISSN 1070-9908. E-ISSN 1558-2361
Grant CEP: GA ČR GA18-05360S
Grant ostatní: GA MŠk(CZ) LO1607
Institucionální podpora: RVO:67985556
Klíčová slova: adaptive image prior * image restoration * variational Bayes
Obor OECD: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Impakt faktor: 3.268, rok: 2018
http://library.utia.cas.cz/separaty/2018/AS/smidl-0490175.pdf

Image restoration is typically defined as an ill-posed problem which has to be regularized to obtain an acceptable solution. In Bayesian interpretation, regularization is equivalent to prior model of the image. An added value of Bayesian point of view is the ability to form a hierarchical model and estimate the hyper-parameters of the prior from the data. Many prior models are available, usually based on automatic relevance determination principle applied to the transformed image. However, the transformation (the most common is a differential operator) is assumed to be known. In this paper, we propose to relax this assumption and estimate the image transformation from the data. The resulting algorithm is analytically tractable using the Variational Bayes method. Properties of the new prior are demonstrated on the problem of image super-resolution.

Trvalý link: http://hdl.handle.net/11104/0284543