Blind deconvolution of images with model discrepancies using maximum a posteriori estimation with heavy-tailed priors

0450671 - ÚTIA 2016 RIV US eng C - Conference Paper (international conference)
Kotera, Jan - Šroubek, Filip
Blind deconvolution of images with model discrepancies using maximum a posteriori estimation with heavy-tailed priors.
Digital Photography XI. Bellingham: SPIE-IS&T, 2015, 94040B. SPIE Proceedings, 9404. ISBN 978-1-62841-494-3. ISSN 0277-786X.
[Digital Photography and Mobile Imaging XI. San Francisco (US), 09.02.2015-10.02.2015]
Grant - others:GA UK(CZ) 938213/2013; GA AV ČR(CZ) M100751201
Institutional support: RVO:67985556
Keywords : image blind deconvolution * image deblurring * blur estimation * pixel saturation * boundary artifacts
Subject RIV: JD - Computer Applications, Robotics
http://library.utia.cas.cz/separaty/2015/ZOI/kotera-0450671.pdf

Single image blind deconvolution aims to estimate the unknown blur from a single observed blurred image and recover the original sharp image. Such task is severely ill-posed and typical approaches involve some heuristic or other steps without clear mathematical explanation to arrive at an acceptable solution. We show that a straight- forward maximum a posteriori estimation incorporating sparse priors and mechanism to deal with boundary artifacts, combined with an efficient numerical method can produce results which compete with or outperform much more complicated state-of-the-art methods. Our method is naturally extended to deal with overexposure in low-light photography, where linear blurring model is violated.
Permanent Link: http://hdl.handle.net/11104/0251952