Title :
An application of the bivariate pseudolikelihood to texture synthesis
Author :
Acuna, Carmen Olga
Author_Institution :
Dept. of Math., Bucknell Univ., Lewisburg, PA, USA
Abstract :
Texture is envisioned as a realization of a Markov random field defined on a square lattice. The model used incorporates both local and global interactions. A modification of the bivariate pseudolikelihood method of parameter estimation is proposed, and the convergence properties of the resulting estimators are examined. The estimated parameters are used to generate textures which are visually compared to those originating from the actual model
Keywords :
Markov processes; convergence of numerical methods; image texture; lattice theory; parameter estimation; Markov random field; bivariate pseudolikelihood; convergence properties; global interactions; local interactions; parameter estimation; square lattice; stochastic model; texture synthesis; Brightness; Histograms; Image segmentation; Lattices; Markov random fields; Mathematics; Parameter estimation; Pixel; Random processes; Stochastic processes;
Conference_Titel :
Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference
Conference_Location :
Austin, TX
Print_ISBN :
0-8186-6952-7
DOI :
10.1109/ICIP.1994.413770
