Self noise and contrast controlled thinning of gray images

Abstract : Homotopic grayscale thinning leads to " over-connected skeleton " when applied on noisy images. One way to avoid this phenomenon is the parametric thinning. It consists in relaxing the initial constraint by lowering low contrast crests, peaks and ends, according to a manually selected parameter and under the constraint of ascendant gray level processing. We propose to control this parameter by considering the lowering decision in a statistical framework of hypothesis test under the assumption of an additive Gaussian noise. A unitary hypothesis test based on the minimum test statistic is used for the elimination of peaks and noise related extremities, while a fusion of multiple tests is required for the insignificant crest lowering decision. This leads to a local adjustment and a standardization of the parametric thinning process that depends only on the chosen significance level of the test.
Type de document :
Article dans une revue
Pattern Recognition, Elsevier, 2016, 57, pp.97 - 114. 〈10.1016/j.patcog.2016.03.033〉
Liste complète des métadonnées

Littérature citée [50 références]  Voir  Masquer  Télécharger
Contributeur : Rabaa Youssef <>
Soumis le : lundi 16 janvier 2017 - 16:34:55
Dernière modification le : jeudi 7 février 2019 - 17:11:52
Document(s) archivé(s) le : lundi 17 avril 2017 - 15:24:33


Fichiers produits par l'(les) auteur(s)




Rabaa Youssef, Sylvie Sevestre-Ghalila, Anne Ricordeau, Amel Benazza. Self noise and contrast controlled thinning of gray images. Pattern Recognition, Elsevier, 2016, 57, pp.97 - 114. 〈10.1016/j.patcog.2016.03.033〉. 〈hal-01436709〉



Consultations de la notice


Téléchargements de fichiers