Autonomous algorithms for image restoration - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Reports Year : 1994

Autonomous algorithms for image restoration


We describe a general theoretical framework for algorithms that adaptively tune all their parameters during the restoration of a noisy image. The adaptation procedure is based on a mean field approach which is known as ``Deterministic Annealing'', and is reminiscent of the ``Deterministic Bolzmann Machiné'. The algorithm is less time consuming in comparison with its simulated annealing alternative. We apply the theory to several architectures and compare their performances.


Other [cs.OH]
Fichier principal
Vignette du fichier
RR-2405.pdf (234.67 Ko) Télécharger le fichier

Dates and versions

inria-00074270 , version 1 (24-05-2006)


  • HAL Id : inria-00074270 , version 1


Meir Griniasty. Autonomous algorithms for image restoration. RR-2405, INRIA. 1994. ⟨inria-00074270⟩
76 View
56 Download


Gmail Facebook Twitter LinkedIn More