Parameter-Free Restoration Algorithms for Two Classes of Binary MRF Images Degraded by Flip-Flap Noises

Bing ZHANG  Mehdi N. SHIRAZI  Hideki NODA  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E80-A   No.10   pp.2022-2031
Publication Date: 1997/10/25
Online ISSN: 
Print ISSN: 0916-8508
Type of Manuscript: PAPER
Category: Image Theory
Bayesian estimation,  sensitivity analysis,  Markov random fields,  Iterated Conditional Modes,  

Full Text: PDF>>
Buy this Article

The problem of restoring binary (black and white) images degraded by color-dependent flip-flap noises is considered. The real image is modeled by a Markov Random Field (MRF). The Iterated Conditional Modes (ICM) algorithm is adopted. It is shown that under certain conditions the ICM algorithm is insensitive to the MRF image model and noise parameters. Using this property, we propose a parameter-free restoration algorithm which does not require the estimations of the image model and noise parameters and thus can be implemented fully in parallel. The effectiveness of the proposed algorithm is shown through applying the algorithm to degraded hand-drawn and synthetic images.