Markov Random Field Texture Models (original) (raw)
2000, IEEE Transactions on Pattern Analysis and Machine Intelligence
We consider a texture to be a stochastic, possibly periodic, two-dimensional image field. A texture model is a mathematical procedure capable of producing and describing a textured image. We explore the use of Markov random fields as texture models. The binomial model, where each point in the texture has a binomial distribution with parameter controlled by its neighbors and ``number of tries'' equal to the number of gray levels, was taken to be the basic model for the analysis. A method of generating samples from the binomial model is given, followed by a theoretical and practical analysis of the method's convergence. Examples show how the parameters of the Markov random field control the strength and direction of the clustering in the image. The power of the binomial model to produce blurry, sharp, line-like, and blob-like textures is demonstrated. Natural texture samples were digitized and their parameters were estimated under the Markov random field model. A hypothesis test was used for an objective assessment of goodness-of-fit under the Markov random field model. Overall, microtextures fit the model well. The estimated parameters of the natural textures were used as input to the generation procedure. The synthetic microtextures closely resembled their real counterparts, while the regular and inhomogeneous textures did not.
Related papers
A Nonparametric Multiscale Markov Random Field Model For Synthesizing Natural Textures
In this paper we present a non-causal, non-parametric, multiscale, Markov random eld (MRF) texture model. The model is capable of capturing the characteristics of and syn- thesising a wide variety of textures, varying from the highly structured to the stochastic. We introduce a novel multi- scale texture synthesis algorithm that allows us to use large neighbourhood systems to model some complex natural tex- tures. As an added advantage of using the novel multi- scale texture synthesis algorithm, phase discontinuities in the synthetic textures are reduced. Finally we show how the high dimensional representation of the texture may be modelled with lower dimensional statistics without compro- mising the integrity of the representation. The power of our modelling technique is evident in that only a small training image is required to derive respectable results even when the texture contains long range characteristics.
A new stochastic image model based on Markov random fields and its application to texture modeling
2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2011
Stochastic image modeling based on conventional Markov random fields is extensively discussed in the literature. A new stochastic image model based on Markov random fields is introduced in this paper which overcomes the shortcomings of the conventional models easing the computation of the joint density function of images. As an application, this model is used to generate texture patterns. The lower computational complexity and easily controllable parameters of the model makes it a more useful model as compared to the conventional Markov random field-based models.
The use of Markov Random Fields as models of texture
Computer Graphics and Image Processing, 1980
... COMPUTER GRAPHICS AND IMAGE PROCESSING 12, 357-370 (1980) The Use of Markov Random Fields as Models of Texture MARTIN HASSNER AND JACK SKLANSKY School of Engineering ... Thus a first-order MRF is determined by three statistical param-eters. ...
Texture Synthesis via a Non-parametric Markov Random Field
2000
In this paper we present a non-causal non-parametric multiscale Markov random field (MRF) texture model that is capable of synthesising a wide variety of textures. The textures that this model is capable of synthesising vary from the highly structured to the stochastic type and include those found in the Brodatz album of textures. The texture model uses Parzen estimation to estimate the conditional probability density function that defines the MRF. For texture synthesis we introduce a novel multiscale approach. We show that these two facets of the model give the ability to model textures requiring large neighbourhood systems to incorporate high order statistical properties of the texture.
Parameter Estimation of Markov Random Field Model of Image Textures
The issue of image feature selection for textured image segmentation is addressed in this paper. It is pointed out that the popular coocurrence matrix statistical method of feature extraction may not be the most efficient one for this task. Usefulness of Random Markov Field parameters as image features is postulated and investigated by means of a numerical analysis. It is found out that MRF parameters can be useful for image segmentation provided they are estimated in an image window of a properly large size. Moreover, the number of features necessary for successful texture classification can be reduced by using the MRFs rather then statistical features, for a wide class of textured images.
Similarity measures for binary and gray level Markov Random Field textures
Lecture Notes in Computer Science, 1997
In this study a new set of texture measures, namely, Clique Length and its moments are introduced. These measures are det~ned employing new concepts which agrees with the human visual system. The simulation experiments are performed on binary and gray level MRF texture alphabet to quantify the data by the k th moment of Clique Length. Experimental results indicate that the introduced measures identify the visually similar textures much better than the mathematical distance measures.
Texture analysis by accurate identification of simple Markovian models
2005
A more accurate identification (estimation of parameters) of simple Markov-Gibbs random field models of images results in a better segmentation of specific multimodal images and realistic synthesis of some types of natural textures. Identification algorithms for segmentation are based in part on a novel modification of an unsupervised learning algorithm published first in ���Cybernetics and Systems Analysis���(���Kibernetika i Sistemnyi Analiz���) almost four decades ago.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.