Unsupervised amplitude and texture based classification of SAR images with multinomial latent model - Archive ouverte HAL Accéder directement au contenu
Rapport Année : 2012

Unsupervised amplitude and texture based classification of SAR images with multinomial latent model

Koray Kayabol
  • Fonction : Auteur
  • PersonId : 900816
Josiane Zerubia
  • Fonction : Auteur
  • PersonId : 833424

Résumé

We combine both amplitude and texture statistics of the Synthetic Aperture Radar (SAR) images for classification purpose. We use Nakagami density to model the class amplitudes and a non-Gaussian Markov Random Field (MRF) texture model with t-distributed regression error to model the textures of the classes. A non-stationary Multinomial Logistic (MnL) latent class label model is used as a mixture density to obtain spatially smooth class segments. The Classification Expectation-Maximization (CEM) algorithm is performed to estimate the class parameters and to classify the pixels. We resort to Integrated Classification Likelihood (ICL) criterion to determine the number of classes in the model. We obtained some classification results of water, land and urban areas in both supervised and unsupervised cases on TerraSAR-X, as well as COSMO-SkyMed data.
Fichier principal
Vignette du fichier
RR-7700v2.pdf (2.9 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00612491 , version 1 (29-07-2011)
hal-00612491 , version 2 (02-05-2012)

Identifiants

  • HAL Id : hal-00612491 , version 2

Citer

Koray Kayabol, Josiane Zerubia. Unsupervised amplitude and texture based classification of SAR images with multinomial latent model. 2012. ⟨hal-00612491v2⟩
292 Consultations
221 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More