Presentation | 2008-03-12 Image Processing by using the EM algorithm and the belief propagation Kei INOUE, Muneki YASUDA, Kazuyuki TANAKA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Markov random fields in image processing includehyperparameters to estimate from given data. We introduce a method to estimate hyperparameters by combining EM algorithm with belief propagation which is familiar computational method in the statistical learning theory. The prior probabilistic model is assumed to be the Q-Ising model and we adopt the symmetric channel and the additive white Gaussian noise as degradation process. In the method, the belief propagation procedures are stopped in a finite number of iterations. We check the influence of these hyperparameter estimations. The obtained results show that the deference between the performance of our proposed method and the one of the conventional EM algorithm with belief propagation is very small even in the case where the number of iterations in the belief propagation is restricted to one. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Probabilistic information processing / Statistical learning / Markov random field / EM algorithm / Belief propagation |
Paper # | NC2007-118 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2008/3/5(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Image Processing by using the EM algorithm and the belief propagation |
Sub Title (in English) | |
Keyword(1) | Probabilistic information processing |
Keyword(2) | Statistical learning |
Keyword(3) | Markov random field |
Keyword(4) | EM algorithm |
Keyword(5) | Belief propagation |
1st Author's Name | Kei INOUE |
1st Author's Affiliation | Graduate School of Information Sciences, Tohoku University() |
2nd Author's Name | Muneki YASUDA |
2nd Author's Affiliation | Graduate School of Information Sciences, Tohoku University |
3rd Author's Name | Kazuyuki TANAKA |
3rd Author's Affiliation | Graduate School of Information Sciences, Tohoku University |
Date | 2008-03-12 |
Paper # | NC2007-118 |
Volume (vol) | vol.107 |
Number (no) | 542 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |