Title :
A multiresolution EM algorithm for unsupervised image classification
Author :
Laferte, J.-M. ; Heitz, F. ; Perez, Pablo
Author_Institution :
IRISA-INRIA, Rennes I Univ.
Abstract :
Using the causal Markov model defined on a quadtree we derive a multiresolution EM algorithm for unsupervised image classification. This algorithm is an efficient alternative to expensive or approximate EM algorithms associated with Markov random fields (MRFs). We show on synthetic and real images that our algorithm also provides good or even better results than those obtained by spatial MRF models
Keywords :
hidden Markov models; image classification; iterative methods; maximum likelihood estimation; quadtrees; causal Markov model; expectation-maximization algorithm; iterative methods; maximum likelihood estimation; multiresolution EM algorithm; parameter estimation; quadtree; unsupervised image classification; Classification algorithms; Computer vision; Hidden Markov models; Image classification; Image resolution; Iterative algorithms; Markov random fields; Maximum likelihood estimation; Spatial resolution; Stochastic processes;
Conference_Titel :
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
Conference_Location :
Vienna
Print_ISBN :
0-8186-7282-X
DOI :
10.1109/ICPR.1996.547196
