Skip to Main content Skip to Navigation
New interface
Journal articles

Caractérisation et classification de textures sur images naturelles

Abstract : The existing texture classification methods are generally based on a parameter extraction stage followed by a classifier stage . Using this kind of method,for an operational application requires to take into account the risk of classes mixture in the parameters space . We propose to take profit of Gagalowicz conjecture in order ta minimise this risk . The conjecture provides us with a set of parameters which totally describe the texture. We show that a connectionnist classifier is able to deal efficiently with these parameters.
Document type :
Journal articles
Complete list of metadata
Contributor : Gilles Burel Connect in order to contact the contributor
Submitted on : Monday, May 10, 2021 - 2:34:07 PM
Last modification on : Monday, November 22, 2021 - 6:42:03 PM


  • HAL Id : hal-03222749, version 1



Gilles Burel, Jean-Yves Catros, Hugues Henocq. Caractérisation et classification de textures sur images naturelles. Traitement du Signal, 1992, 9 (1), pp.33-43. ⟨hal-03222749⟩



Record views