Multiresolution image classification by hierarchical modeling with two-dimensional hidden Markov models

This paper treats a multiresolution hidden Markov model for classifying ima ges. Each image is represented by feature vectors at several resolutions, w hich are statistically dependent as modeled by the underlying state process , a multiscale Markov mesh. Unknowns in the model are estimated by maximum likelihood, in particular by employing the expectation-maximization algorit hm. An image is classified by finding the optimal set of states with maximu m a posteriori probability. States are then mapped into classes. The multir esolution model enables multiscale information about context to be incorpor ated into classification. Suboptimal algorithms based on the model provide progressive classification that is much faster than the algorithm based on single-resolution hidden Markov models.