Reconstruction of gray-scale images

DOI10.1023/A:1013762722096MaRDI QIDQ5959765zbMATH OpenOpenAlexFDO

Authors Pablo A. Ferrari, Marco Dimas Gubitoso, E. Jordão Neves

Publication date 11 April 2002

Published in Methodology and Computing in Applied Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0003098

zbMATH Keywords

fast algorithmsmaximum a posteriori estimatormulticolor reconstruction

Mathematics Subject Classification ID

Directional data; spatial statistics (62H11) Random fields; image analysis (62M40) Computing methodologies for image processing (68U10)

Abstract: Reconstruction of images corrupted by noise is an important problem in Image Analysis. In the standard Bayesian approach the unknown original image is assumed to be a realization of a Markov random field on a finite two dimensional finite region. This image is degraded by some noise, which acts independently in each site of and has the same distribution on all sites. The reconstructed image is the maximum a posteriori (MAP) estimator. When the number of colors is bigger than two the standard a priori measure is the Potts model, but with this choice reconstruction algorithms need exponential growing time in the number of pixels. We propose a different a priori measure that allows to compute the MAP estimator in polinomial time. The algorithm binary decomposes the intensity of each color and reconstructs each component, reducing the problem to the computation of the two-color MAP estimator. The latter is done in polinomial time solving a min-cut max-flow problem in a binary graph as proposed by Greig, Porteous and Seheult (1989).

Recommendations

• scientific article; zbMATH DE number 795290
• Model distortions in Bayesian MAP reconstruction
• scientific article; zbMATH DE number 427518
• scientific article; zbMATH DE number 1551780
• scientific article; zbMATH DE number 3986503

Cited in

(13)

• Probabilistic linear models for multiresolution estimation in gray-level images
• Model distortions in Bayesian MAP reconstruction
• Exact distribution of edge-preserving MAP estimators for linear signal models with Gaussian measurement noise
• Reconstruction of Bicolored Images
• Gray levels can improve the performance of binary image digitizers
• scientific article; zbMATH DE number 427518 (Why is no real title available?)
• scientific article; zbMATH DE number 4174807 (Why is no real title available?)
• scientific article; zbMATH DE number 4045083 (Why is no real title available?)
• scientific article; zbMATH DE number 1551784 (Why is no real title available?)
• Estimation of the Number of True Gray Levels, Their Values, and Relative Frequencies in a Noisy Image
• scientific article; zbMATH DE number 1423109 (Why is no real title available?)
• Computational Science and Its Applications – ICCSA 2004
• Network flow optimization for restoration of images

This page was built for publication: Reconstruction of gray-scale images

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5959765)