Skip to Main content Skip to Navigation
Conference papers

Restauration d'image par une approche bayésienne semi non supervisée et le mélange de gaussienne

Abstract : Convex image restoration is a major field in inverse problems, and problem is often addressed by hand-tuning hyper-parameters. We propose a contribution about a Bayesian approach where a convex field is constructed via Location Mixture of Gaussian and the estimator computed with a fast MCMC algorithm. Main contributions are the use of several operator avoiding crosslike artifacts and a new algorithm to simulate prior mean. Results, in comparison to standard supervised results, have equivalent quality in a quasi-unsupervised approach and go with uncertainty quantification.
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal-centralesupelec.archives-ouvertes.fr/hal-01637848
Contributor : François Orieux <>
Submitted on : Thursday, March 5, 2020 - 11:10:05 AM
Last modification on : Wednesday, April 8, 2020 - 4:03:27 PM
Document(s) archivé(s) le : Saturday, June 6, 2020 - 1:32:40 PM

File

gretsifr.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01637848, version 1

Citation

François Orieux, Raphael Chinchilla. Restauration d'image par une approche bayésienne semi non supervisée et le mélange de gaussienne. 26eme Colloque GRETSI Traitement du Signal & des Images, GRETSI 2017, Sep 2017, Juan-les-Pins, France. ⟨hal-01637848⟩

Share

Metrics

Record views

201

Files downloads

6