K. Kayabol, V. Krylov, J. Zerubia.
"Unsupervised Classification of SAR Images using Hierarchical Agglomeration and EM",
MUSCLE International Workshop on Computational Intelligence for Multimedia Understanding,
Springer LNCS 7252, Pp. 54-65, Pisa (Italy), December 13-15, 2011.
[link] [pdf]
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.
Bibtex
@INPROCEEDINGS{KayabolLNCS11,
author = {Kayabol, K. and Krylov, V. and Zerubia, J.},
title = {Unsupervised Classification of {SAR} Images using Hierarchical Agglomeration and {EM}},
year = {2011},
booktitle = {Springer LNCS},
volume = {7252},
address = {Pisa, Italy},
pages = {54--65}
}