Textured image segmentation using multiresolution Markov Random Fields and a two-component texture model

In this paper we propose a multiresolution Markov Random Field (MMRF) model for segmenting textured images. The Multiresolution Fourier Transform (MFT) is used to provide a set of spatially localised texture descriptors, which are based on a two-component model of texture, in which one component is a deformation, representing the structural or deterministic elements and the other is a stochastic one. Stochastic relaxation labelling is adopted to maximise the likelihood and assign the class label with highest probability to the block (site) being visited. Class information is propagated from low spatial resolution to high spatial resolution, via appropriate modifications to the interaction energies defining the field, to minimise class-position uncertainty. Experiments on the segmentation of natural textures are used to show the potential of the method.