Decision rules for choice of neighbors in random field models of images

R. L. Kashyap, R. Chellappa, Narendra Ahuja

Research output: Contribution to journalArticle

Abstract

Random field models have many applications in image processing and analysis. The mainconcern of this paper is to design a decision rule for fitting an appropriate random field model to a given image. We assume that the given image is a particular realization of a homogenous Gaussian discrete random field. We represent the underlying random field by a set of parametric models representing the spatial dependence. Using spectral representations of the random field and standard Bayesian methods, we develop a decision rule for choosing an appropriate model from a class of such models. We discuss the relevance of the theory developed in this paper for applications in image modeling and texture characterization.

Original languageEnglish (US)
Pages (from-to)301-318
Number of pages18
JournalComputer Graphics and Image Processing
Volume15
Issue number4
DOIs
StatePublished - Apr 1981

Fingerprint

image processing
image analysis
texture
decision
modeling
method

ASJC Scopus subject areas

  • Environmental Science(all)
  • Earth and Planetary Sciences(all)

Cite this

Decision rules for choice of neighbors in random field models of images. / Kashyap, R. L.; Chellappa, R.; Ahuja, Narendra.

In: Computer Graphics and Image Processing, Vol. 15, No. 4, 04.1981, p. 301-318.

Research output: Contribution to journalArticle

@article{a6a449108a7443b4917972a059586e7c,
title = "Decision rules for choice of neighbors in random field models of images",
abstract = "Random field models have many applications in image processing and analysis. The mainconcern of this paper is to design a decision rule for fitting an appropriate random field model to a given image. We assume that the given image is a particular realization of a homogenous Gaussian discrete random field. We represent the underlying random field by a set of parametric models representing the spatial dependence. Using spectral representations of the random field and standard Bayesian methods, we develop a decision rule for choosing an appropriate model from a class of such models. We discuss the relevance of the theory developed in this paper for applications in image modeling and texture characterization.",
author = "Kashyap, {R. L.} and R. Chellappa and Narendra Ahuja",
year = "1981",
month = "4",
doi = "10.1016/S0146-664X(81)80014-7",
language = "English (US)",
volume = "15",
pages = "301--318",
journal = "Computer Graphics and Image Processing",
issn = "0146-664X",
publisher = "Academic Press Inc.",
number = "4",

}

TY - JOUR

T1 - Decision rules for choice of neighbors in random field models of images

AU - Kashyap, R. L.

AU - Chellappa, R.

AU - Ahuja, Narendra

PY - 1981/4

Y1 - 1981/4

N2 - Random field models have many applications in image processing and analysis. The mainconcern of this paper is to design a decision rule for fitting an appropriate random field model to a given image. We assume that the given image is a particular realization of a homogenous Gaussian discrete random field. We represent the underlying random field by a set of parametric models representing the spatial dependence. Using spectral representations of the random field and standard Bayesian methods, we develop a decision rule for choosing an appropriate model from a class of such models. We discuss the relevance of the theory developed in this paper for applications in image modeling and texture characterization.

AB - Random field models have many applications in image processing and analysis. The mainconcern of this paper is to design a decision rule for fitting an appropriate random field model to a given image. We assume that the given image is a particular realization of a homogenous Gaussian discrete random field. We represent the underlying random field by a set of parametric models representing the spatial dependence. Using spectral representations of the random field and standard Bayesian methods, we develop a decision rule for choosing an appropriate model from a class of such models. We discuss the relevance of the theory developed in this paper for applications in image modeling and texture characterization.

UR - http://www.scopus.com/inward/record.url?scp=0019728198&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0019728198&partnerID=8YFLogxK

U2 - 10.1016/S0146-664X(81)80014-7

DO - 10.1016/S0146-664X(81)80014-7

M3 - Article

AN - SCOPUS:0019728198

VL - 15

SP - 301

EP - 318

JO - Computer Graphics and Image Processing

JF - Computer Graphics and Image Processing

SN - 0146-664X

IS - 4

ER -