Physics – Condensed Matter – Disordered Systems and Neural Networks
Scientific paper
2008-07-21
Physics
Condensed Matter
Disordered Systems and Neural Networks
13pages, 12figures, using elsart.cls
Scientific paper
On the basis of statistical mechanics of the Q-Ising model, we formulate the Bayesian inference to the problem of inverse halftoning, which is the inverse process of representing gray-scales in images by means of black and white dots. Using Monte Carlo simulations, we investigate statistical properties of the inverse process, especially, we reveal the condition of the Bayes-optimal solution for which the mean-square error takes its minimum. The numerical result is qualitatively confirmed by analysis of the infinite-range model. As demonstrations of our approach, we apply the method to retrieve a grayscale image, such as standard image `Lenna', from the halftoned version. We find that the Bayes-optimal solution gives a fine restored grayscale image which is very close to the original.
Inoue Jun'ichi
Okada Masato
Saika Yohei
Tanaka Hiroyuki
No associations
LandOfFree
Bayes-optimal inverse halftoning and statistical mechanics of the Q-Ising model does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Bayes-optimal inverse halftoning and statistical mechanics of the Q-Ising model, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bayes-optimal inverse halftoning and statistical mechanics of the Q-Ising model will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-71597