Understanding image priors in blind deconvolution

0434275 - ÚTIA 2015 RIV US eng C - Conference Paper (international conference)
Šroubek, Filip - Šmídl, Václav - Kotera, Jan
Understanding image priors in blind deconvolution.
2014 IEEE International Conference on Image Processing. USA: IEEE, 2014, s. 4492-4496. ISBN 978-1-4799-5751-4.
[2014 IEEE International Conference on Image Processing. Paris (FR), 27.10.2014-30.10.2014]
R&D Projects: GA ČR GA13-29225S
Grant - others:Grantová agentura UK(CZ) GAUK 938213
Institutional support: RVO:67985556
Keywords : blind deconvolution * variational Bayes * automatic relevance determination
Subject RIV: JD - Computer Applications, Robotics
http://library.utia.cas.cz/separaty/2014/ZOI/sroubek-0434275.pdf

Removing blurs from a single degraded image without any knowledge of the blur kernel is an ill-posed blind deconvolution problem. Proper estimators together with correct image priors play a fundamental role in accurate blind deconvolution. We demonstrate a superior performance of the variational Bayesian estimator and discuss suitability of automatic relevance determination distributions as image priors. Restoration of real photos blurred by out-of-focus and motion blur, and comparison with a state-of-the-art method is provided.
Permanent Link: http://hdl.handle.net/11104/0241853