bibtype	J - Journal Article
ARLID	0376080
utime	20240111140815.6
mtime	20120511235959.9
WOS	000302181800022
SCOPUS	84859087283
DOI	10.1109/TIP.2011.2175740
title (primary) (eng)	Robust Multichannel Blind Deconvolution via Fast Alternating Minimization
specification	page_count 14 s.
serial	ARLID cav_un_epca*0253235 ISSN 1057-7149 title IEEE Transactions on Image Processing volume_id 21 volume 4 (2012) page_num 1687-1700 publisher name Institute of Electrical and Electronics Engineers
keyword	blind deconvolution
keyword	augmented Lagrangian
keyword	sparse representation
author (primary)	ARLID cav_un_auth*0101209 name1 Šroubek name2 Filip full_dept (cz) Zpracování obrazové informace full_dept (eng) Department of Image Processing department (cz) ZOI department (eng) ZOI institution UTIA-B full_dept Department of Image Processing fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
author	ARLID cav_un_auth*0274481 name1 Milanfar name2 P. country US
source	source_type pdf url http://library.utia.cas.cz/separaty/2012/ZOI/sroubek-0376080.pdf source_size 2.5MB
cas_special	project project_id 1M0572 agency GA MŠk ARLID cav_un_auth*0001814 project project_id GAP103/11/1552 agency GA ČR ARLID cav_un_auth*0273618 project project_id VG20102013064 agency GA MV ARLID cav_un_auth*0273814 research CEZ:AV0Z10750506 abstract (eng) Blind deconvolution, which comprises simultaneous blur and image estimations, is a strongly ill-posed problem. It is by now well known that if multiple images of the same scene are acquired, this multichannel (MC) blind deconvolution problem is better posed and allows blur estimation directly fromthe degraded images.We improve theMC idea by adding robustness to noise and stability in the case of large blurs or if the blur size is vastly overestimated. We formulate blind deconvolution as an–regularized optimization problemand seek a solution by alternately optimizing with respect to the image and with respect to blurs. Each optimization step is converted to a constrained problem by variable splitting and then is addressed with an augmented Lagrangian method, which permits simple and fast implementation in the Fourier domain. The rapid convergence of the proposed method is illustrated on synthetically blurred data. Applicability is also demonstrated on the deconvolution of real photos taken by a digital camera. reportyear 2013 RIV JD mrcbC52 4 A 4a 20231122135022.2 permalink http://hdl.handle.net/11104/0208579 mrcbT16-e COMPUTERSCIENCEARTIFICIALINTELLIGENCE|ENGINEERINGELECTRICALELECTRONIC mrcbT16-f 4.017 mrcbT16-g 0.407 mrcbT16-h 7.6 mrcbT16-i 0.03545 mrcbT16-j 1.642 mrcbT16-k 13814 mrcbT16-l 408 mrcbT16-q 156 mrcbT16-s 1.521 mrcbT16-y 35.94 mrcbT16-x 5.36 mrcbT16-4 Q1 mrcbT16-B 95.396 mrcbT16-C 91.170 mrcbT16-D Q1* mrcbT16-E Q1* arlyear 2012 mrcbTft \nSoubory v repozitáři: sroubek-0376080.pdf mrcbU14 84859087283 SCOPUS mrcbU34 000302181800022 WOS mrcbU56 pdf 2.5MB mrcbU63 cav_un_epca*0253235 IEEE Transactions on Image Processing 1057-7149 1941-0042 Roč. 21 č. 4 2012 1687 1700 Institute of Electrical and Electronics Engineers