Skip to Main content Skip to Navigation
Theses

Flou et quantification dans les images numériques

Abstract : The first part of this thesis deals with the dequantization of digital images. We present a dequantization scheme that enhances the statistical properties of the gradient feld. We apply this method to the meaningful alignments detection developed by Agnès Desolneux. The second part presents a method to evaluate the blur in natural images. We use the morphological scale space to derive an accurate local evaluation of the blur in the image. We do not make any assumption on the type of the blur kernel.
Document type :
Theses
Complete list of metadata

Cited literature [43 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00133427
Contributor : Ens Cachan Bibliothèque <>
Submitted on : Monday, February 26, 2007 - 12:24:31 PM
Last modification on : Thursday, April 15, 2021 - 3:31:33 AM
Long-term archiving on: : Tuesday, April 6, 2010 - 9:34:08 PM

Identifiers

  • HAL Id : tel-00133427, version 1

Citation

Saïd Ladjal. Flou et quantification dans les images numériques. Mathématiques [math]. École normale supérieure de Cachan - ENS Cachan, 2005. Français. ⟨tel-00133427⟩

Share

Metrics

Record views

594

Files downloads

6074