Steganalysis by Ensemble Classifiers with Boosting by Regression, and Post-Selection of Features

Marc Chaumont 1 Sarra Kouider 1
1 ICAR - Image & Interaction
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : In this paper we extend the state-of-the-art steganalysis tool developed by Kodovsky and Fridrich: the Kodovsky's ensem-ble classifiers. We propose to boost the weak classifiers com-posing the Kodovsky classifier. For this, we minimize the probability of error thanks to a regression approach of low complexity. We also propose a post-selection of features, achieved after the learning step of all the weak classifiers. For each weak classifier, we identify a subset of features reducing the probability of error. Both proposals are of neg-ligeable complexity compared to the complexity of the Kodovsky classifier. Moreover, these two proposals significantly increase the performance of classification.
Type de document :
Communication dans un congrès
ICIP: International Conference on Image Processing, Sep 2012, Orlando, FL, United States. 19th IEEE International Conference on Image Processing, pp.1133-1136, 2012, 〈http://icip2012.com/〉. 〈10.1109/ICIP.2012.6467064〉
Liste complète des métadonnées

Littérature citée [12 références]  Voir  Masquer  Télécharger

https://hal-lirmm.ccsd.cnrs.fr/lirmm-00838995
Contributeur : Marc Chaumont <>
Soumis le : lundi 1 juillet 2013 - 18:50:25
Dernière modification le : jeudi 24 mai 2018 - 15:59:23
Document(s) archivé(s) le : mercredi 2 octobre 2013 - 03:05:09

Identifiants

Collections

Citation

Marc Chaumont, Sarra Kouider. Steganalysis by Ensemble Classifiers with Boosting by Regression, and Post-Selection of Features. ICIP: International Conference on Image Processing, Sep 2012, Orlando, FL, United States. 19th IEEE International Conference on Image Processing, pp.1133-1136, 2012, 〈http://icip2012.com/〉. 〈10.1109/ICIP.2012.6467064〉. 〈lirmm-00838995〉

Partager

Métriques

Consultations de la notice

159

Téléchargements de fichiers

487