A COMBINED MARKOV RANDOM-FIELD AND WAVE-PACKET TRANSFORM-BASED APPROACH FOR IMAGE SEGMENTATION

The present work formulates a novel segmentation algorithm which combi nes the use of Markov random field models for image-modeling with the use of the discrete wave-packet transform for image analysis. Image se gmentations are derived and refined at a sequence of resolution levels , using as data selected wave-packet transform images or ''channels.'' The described segmentation algorithm is compared with non-multiresolu tion Markov random field-based image segmentation algorithms in the co ntest of synthetic image example problems, and found to be both signif icantly more efficient and effective.