Segmentation using Wavelet - domain Classi

We introduce a new image texture segmentation algorithm, HMTseg, based on wavelet-domain hidden Markov tree (HMT) models. The HMT model is a tree-structured probabilistic graph that captures the statistical properties of wavelet coeecients. Since the HMT is particularly well suited to images containing singularities (edges and ridges), it provides provides a good classiier for textures. Utilizing the inherent tree structure of the wavelet HMT and its fast training and likelihood computation algorithms, we perform multiscale texture classiication at various scales. We then fuse these multiscale classiications using a Bayesian probabilistic graph to obtain a reliable nal segmentation. Since HMTseg works on the wavelet transform of the image, it can directly segment wavelet-compressed images, without the need for decompression. We demonstrate the performance of HMTseg with synthetic, aerial photo, and document image segmentations.