Title :
Texture analysis with variational hidden Markov trees
Author :
Dasgupta, Nilanjan ; Carin, Lawrence
Author_Institution :
Dept. of Electr. & Comput. Eng., Duke Univ., Durham, NC, USA
fDate :
6/1/2006 12:00:00 AM
Abstract :
A variational Bayes formulation of the hidden Markov tree (HMT) model is proposed for texture analysis, utilizing a multilevel wavelet decomposition of imagery. The variational method yields an approximation to the full posterior of the HMT parameters. Texture classification is based on the posterior predictive distribution or marginalized evidence, with example results presented.
Keywords :
Bayes methods; hidden Markov models; image classification; image texture; trees (mathematics); wavelet transforms; imagery multilevel wavelet decomposition; porterior predictive distribution; texture analysis; texture classification; variational Bayes formulation; variational hidden Markov trees; Hidden Markov models; Image analysis; Image classification; Image texture analysis; Maximum likelihood estimation; Statistical distributions; Training data; Two dimensional displays; Wavelet analysis; Wavelet coefficients; HMT; Kullback–Leibler divergence; texture classification; variational Bayes;
Journal_Title :
Signal Processing, IEEE Transactions on
DOI :
10.1109/TSP.2006.872588
