Markov / Gibbs modeling : Texture and Temperature
نویسندگان
چکیده
Markov/Gibbs random elds have been used for posing a variety of computer vision and image processing problems. Many of these problems are then solved using a simulated annealing type of method which involves the varying of the \temperature," a scale parameter for the model. In this paper we analyze the eeect of temperature on random eld texture patterns. We obtain new results relating structure in the texture co-occurrence matrix to temperature. We also show the existence of multiple \transition tempera-tures" which delimit regions of diierent band-width in the co-occurrence matrix, and hence can be used to control pattern formation.
منابع مشابه
Markov/Gibbs texture modeling: aura matrices and temperature effects
We use a new \aura" framework to rewrite the non-linear energy function of a homogeneous anisotropic Markov/Gibbs random eld (MRF) as a linear sum of aura measures. The new formulation relates MRF's to co-occurrence matrices. It also provides a physical interpretation of MRF textures in terms of the mixing and separation of graylevel sets, and in terms of boundary maximization and minimization....
متن کاملTexture Synthesis: From Convolutional RBMs to Efficient Deterministic Algorithms
Probabilistic models of textures should be able to synthesize specific textural structures, prompting the use of filter-based Markov random fields (MRFs) with multi-modal potentials, or of advanced variants of restricted Boltzmann machines (RBMs). However, these complex models have practical problems, such as inefficient inference, or their large number of model parameters. We show how to train...
متن کاملFilters, Random Elds and Maximum Entropy (frame) |towards a Uniied Theory for Texture Modeling
This article presents a statistical theory for texture modeling. This theory combines ltering theory and Markov random eld modeling through the maximum entropy principle , and interprets and clariies many previous concepts and methods for texture analysis and synthesis from a uniied point of view. Our theory characterizes the ensemble of images I with the same texture appearance by a probabilit...
متن کاملFilters , Random elds And Maximum Entropy ( FRAME ) | Towards a Uni ed Theory for Texture
This article presents a statistical theory for texture modeling. This theory combines ltering theory and Markov random eld modeling through the maximum entropy principle, and interprets and clariies many previous concepts and methods for texture analysis and synthesis from a uniied point of view. Our theory characterizes the ensemble of images I with the same texture appearance by a probability...
متن کاملTexture analysis and simulations using Markov random field models
The main goal of texture analysis is to extract useful textural information from an image. The use of Bayesian methods is an approach, which seeks to provide a unified framework in modeling within many different image processes. In this work, spatial behavior of the auto-logistic model in a rectangular lattice would be investigated, concentrating the first-order neighborhood structures. A simpl...
متن کامل