Unsupervised Classification of SAR Images Using Hierarchical Agglomeration and EM

Koray Kayabol 1 Vladimir A. Krylov 1 Josiane Zerubia 1
1 ARIANA - Inverse problems in earth monitoring
CRISAM - Inria Sophia Antipolis - Méditerranée , SIS - Signal, Images et Systèmes
Abstract : We implement an unsupervised classification algorithm for high resolution Synthetic Aperture Radar (SAR) images. The foundation of algorithm is based on Classification Expectation-Maximization (CEM). To get rid of two drawbacks of EM type algorithms, namely the initialization and the model order selection, we combine the CEM algorithm with the hierarchical agglomeration strategy and a model order selection criterion called Integrated Completed Likelihood (ICL). We exploit amplitude statistics in a Finite Mixture Model (FMM), and a Multinomial Logistic (MnL) latent class label model for a mixture density to obtain spatially smooth class segments. We test our algorithm on TerraSAR-X data.
Type de document :
Communication dans un congrès
Emanuele Salerno and A. Enis Çetin and Ovidio Salvetti. Computational Intelligence for Multimedia Understanding MUSCLE 2011, Dec 2011, Pisa, Italy. Springer, 7252, pp.54-65, 2012, Lecture Notes in Computer Science - LNCS; Computational Intelligence for Multimedia Understanding. <10.1007/978-3-642-32436-9_5>


https://hal.inria.fr/hal-00782641
Contributeur : Koray Kayabol <>
Soumis le : jeudi 31 janvier 2013 - 15:43:22
Dernière modification le : mercredi 20 mars 2013 - 17:05:33
Document(s) archivé(s) le : mercredi 1 mai 2013 - 03:55:02

Fichier

MUSCLE11v3.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Koray Kayabol, Vladimir A. Krylov, Josiane Zerubia. Unsupervised Classification of SAR Images Using Hierarchical Agglomeration and EM. Emanuele Salerno and A. Enis Çetin and Ovidio Salvetti. Computational Intelligence for Multimedia Understanding MUSCLE 2011, Dec 2011, Pisa, Italy. Springer, 7252, pp.54-65, 2012, Lecture Notes in Computer Science - LNCS; Computational Intelligence for Multimedia Understanding. <10.1007/978-3-642-32436-9_5>. <hal-00782641>

Partager

Métriques

Consultations de
la notice

167

Téléchargements du document

156