. "0"^^ . . . "22"^^ . "P\u0159edlo\u017Een\u00FD projekt se zab\u00FDv\u00E1 f\u00E1z\u00ED digit\u00E1ln\u00EDch obraz\u016F v p\u0159\u00EDpad\u011B, \u017Ee jednotliv\u00E9 vstupn\u00ED obrazy (kan\u00E1ly)jsou po\u0161kozeny nezn\u00E1m\u00FDmi line\u00E1rn\u00EDmi filtry, geometrick\u00FDm zkreslen\u00EDm a aditivn\u00EDm \u0161umem. F\u00FAz\u00ED zde rozum\u00EDme proces extrakce %22u\u017Eite\u010Dn\u00E9%22 \u010Di %22nepo\u0161kozen\u00E9%22 informace z jednotliv\u00FDch kan\u00E1l\u016F a jej\u00ED integrace do jednoho v\u00FDsledn\u00E9ho obrazu, jeho\u017E kvalita (a t\u00EDm i hodnota pro n\u00E1slednou anal\u00FDzu) je na vy\u0161\u0161\u00ED \u00FArovni ne\u017E je kvalita vstup\u016F. V na\u0161em pojet\u00ED je f\u00FAze prov\u00E1d\u011Bna ve dvou kroc\u00EDch: geometrick\u00E1 registrace vstupn\u00EDch obraz\u016F a v\u00EDcekan\u00E1lov\u00E1 slep\u00E1 dekonvoluce. V prvn\u00ED \u010D\u00E1sti bude na\u0161\u00EDm \u00FAkolem nal\u00E9zt metodu (metody) automatick\u00E9 registrace obraz\u016F degradovan\u00FDch filtrac\u00ED. Tato \u00FAloha nen\u00ED ve sv\u011Bt\u011B vy\u0159e\u0161ena, bylo pouze uk\u00E1z\u00E1no, \u017Ee standardn\u00ED metody vyvinut\u00E9 pro nedegradovan\u00E9 obrazy nelze pou\u017E\u00EDt. P\u0159edpokl\u00E1d\u00E1me, \u017Ee na\u0161e \u0159e\u0161en\u00ED bude zalo\u017Eeno na tzv. kombinovan\u00FDch konvolu\u010Dn\u00EDch invariantech.Ve druh\u00E9 \u010D\u00E1sti projektu chceme navrhnout novou metodu v\u00EDcekan\u00E1lov\u00E9 slep\u00E9 dekonvoluce. N\u011Bkolik m\u00E1lo takov\u00FDch metod bylo v" . "22"^^ . "Neuvedeno."@en . . "Image fusion methods for degraded and incomplete data"@en . . "2008-05-19+02:00"^^ . "The project is concerned with an automatic fusion of digital images. The input images are assumed to be blurred by unknown linear filters, geometrically distorted and corrupted by additive noise. By image fusion we understand the process of extraction ofuseful information content from each individual input frame and its integration into one output image, the quality of which is of higher level than that of the inputs.In this project, we propose to carry out the fusion in two stages: (a) geometric registration of the input frames and (b) multiframe blind deconvolution. Since the current registration techniques were proven to fail when applied to blurred images, we want to develop a new method especially for this purpose. This method is assumed to be based on our recently derived invariants to blurring. In the second stage, we want to propose a new method for multiframe blind deconvolution, which should be more numerically stable and faster than the previously published ones. In this way, the inpu"@en . . . "Metody obrazov\u00E9 f\u00FAze pro po\u0161kozen\u00E1 a ne\u00FApln\u00E1 data" . . . "http://www.isvav.cz/projectDetail.do?rowId=GA102/00/1711"^^ . . . "1"^^ . . "1"^^ . . "GA102/00/1711" . . . "Projekt se zab\u00FDval automatickou f\u00FAz\u00ED digit\u00E1ln\u00EDch obraz\u016F v p\u0159\u00EDpadech, \u017Ee vstupn\u00ED data jsou siln\u011B po\u0161kozena nezn\u00E1m\u00FDm typem rozost\u0159en\u00ED, geometrick\u00FDm zkreslen\u00EDm nebo aditivn\u00EDm \u0161umem, pop\u0159. v\u0161emi t\u011Bmito vlivy sou\u010Dasn\u011B. Teoretick\u00E9 v\u00FDsledky byly aplikov\u00E1ny v as"@cs . . .