Bayesian approach to blind deconvolution based on Dirichlet distributions

Rafael Molina*, Aggelos K. Katsaggelos, Javier Abad, Javier Mateos

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

27 Scopus citations


This paper deals with the simultaneous identification of the blur and the restoration of a noisy and blurred image. We propose the use of Dirichlet distributions to model our prior knowledge about the blurring function together with smoothness constraints on the restored image to solve the blind deconvolution problem. We show that the use of Dirichlet distributions offers a lot of flexibility in incorporating vague or very precise knowledge about the blurring process into the blind deconvolution process. The proposed MAP estimator offers additional flexibility in modeling the original image. Experimental results demonstrate the performance of the proposed algorithm.

Original languageEnglish (US)
Pages (from-to)2809-2812
Number of pages4
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
StatePublished - Jan 1 1997

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics


Dive into the research topics of 'Bayesian approach to blind deconvolution based on Dirichlet distributions'. Together they form a unique fingerprint.

Cite this