Temporal or dynamic textures are video sequences that are spatially repetitive and temporally stationary. Dynamic textures are temporal analogs of the more common spatial texture. They are a family of visual phenomena where the texture elements, or the spatially repeating elements, undergo stochastic motions that are statistically similar. Dynamic textures appear in a vast spectrum of videos, ranging from sequences of moving water, foliage, smoke, and clouds to sequences of swarms of birds, robots, and even humans in crowds. Also, the applications concerning such video sequences are significant and numerous, including surveillance (e.g. monitoring traffic or crowds), detection of the onset of emergencies (e.g. outbreak of fires), and foreground and background separation (e.g. the transfer of a dynamic texture from one environment to another or simply dynamic texture removal). The study of dynamic textures poses numerous challenges, especially for traditional motion models that fail to capture the stochastic nature of dynamic textures. Despite their importance, the study of dynamic textures has just recently attracted the attention of the computer vision community. After modeling dynamic textures, it can be used for dynamic texture synthesis, compression, categorization and segmentation. The main goal of this thesis is dynamic texture segmentation. For dynamic texture segmentation, first using a learning process, model parameters are extracted, then a region-based segmentation technique, using level set methods and generalization of Chan-Vese model in time is designed. This technique can segment dynamic texture videos into regions with different spatiotemporal statistics. In this method the spatial statistics of a dynamic texture is modeled with a set of texture descriptors whose temporal evolution is governed by an AutoRegressive (AR) model. Also different feature vectors were investigated and experimental results showed that directional Walsh Hadamard transform has a high performance to extract texture features. As an application, we used dynamic texture techniques for medical issue segmentation. Due to characteristics of MR images and to heart anatomy, the problem of heart localization and segmentation in MRI is still open. Because the heart is the only moving organ in the images and other organs like chest and liver are static, heart can be considered as a dynamic texture. Modeling heart as a dynamic texture can be used for heart localization and epicardium and endocardium segmentation. A fast and robust algorithm is proposed for heart localization and dynamic texture segmentation techniques are used for epicardial segmentation. Also we proposed a dynamic texture based method for endocardium segmentation. Experimental results are compared to manually segmented results and evaluated by numerical measure. The correct localization rate is 100% and the proposed method for epicardial segmentation achieved encouraging results, accuracy rate 0.98. Keywords: Dynamic texture segmentation, AR model, Directional Walsh Hadamard, Linear Dynamic Systems, Cardiac MRI, endocardium segmentation.