Title :
A study of contextual modeling and texture characterization for multiscale Bayesian segmentation
Author :
Fan, Guoliang ; Song, Xiaomu
Author_Institution :
Sch. of Electr. & Comput. Eng., Oklahoma State Univ., Stillwater, OK, USA
Abstract :
In this paper, we demonstrate that multiscale Bayesian image segmentation can be enhanced by improving both contextual modeling and statistical texture characterization. Firstly, we show a joint multi-context and multiscale approach to achieve more robust contextual modeling by using multiple context models. Secondly, we study statistical texture characterization using wavelet-domain hidden Markov models (HMMs), and in particular, we use an improved HMM, HMT-3S, to obtain more accurate multiscale texture characterization. Experimental results show that both of them play important roles in multiscale Bayesian segmentation.
Keywords :
Bayes methods; hidden Markov models; image segmentation; image texture; wavelet transforms; HMM; HMT-3S; contextual modeling; image segmentation; multiple context models; multiscale Bayesian segmentation; multiscale texture characterization; statistical texture characterization; wavelet-domain hidden Markov models; Bayesian methods; Computational efficiency; Context modeling; Cost function; Discrete wavelet transforms; Hidden Markov models; Image segmentation; Markov random fields; Robustness; Statistics;
Conference_Titel :
Image Processing. 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7622-6
DOI :
10.1109/ICIP.2002.1038967