Template Matching with Noisy Patches: A Contrast-Invariant GLR Test

Abstract : Matching patches from a noisy image to atoms in a dictionary of patches is a key ingredient to many techniques in image processing and computer vision. By representing with a single atom all patches that are identical up to a radiometric transformation, dictionary size can be kept small, thereby retaining good computational efficiency. Identification of the atom in best match with a given noisy patch then requires a contrast-invariant criterion. In the light of detection theory, we propose a new criterion that ensures contrast invariance and robustness to noise. We discuss its theoretical grounding and assess its performance under Gaussian, gamma and Poisson noises.
Type de document :
Communication dans un congrès
European Signal Processing Conference 2013, Sep 2013, Marrakech, Morocco
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-00804483
Contributeur : Charles-Alban Deledalle <>
Soumis le : lundi 25 mars 2013 - 15:51:21
Dernière modification le : mercredi 22 février 2017 - 09:27:59
Document(s) archivé(s) le : mercredi 26 juin 2013 - 04:03:12

Fichiers

paper_hal.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00804483, version 1
  • ARXIV : 1303.6152

Citation

Charles-Alban Deledalle, Loïc Denis, Florence Tupin. Template Matching with Noisy Patches: A Contrast-Invariant GLR Test. European Signal Processing Conference 2013, Sep 2013, Marrakech, Morocco. <hal-00804483>

Partager

Métriques

Consultations de
la notice

404

Téléchargements du document

167