A Bayesian approach to blind deconvolution based on Dirichlet distributions.
Rafael MolinaAggelos K. KatsaggelosJavier AbadJavier MateosPublished in: ICASSP (1997)
Keyphrases
- blind deconvolution
- dirichlet distribution
- prior distribution
- mixture model
- bayesian framework
- total variation
- discrete data
- image deblurring
- map estimation
- point spread function
- blurred images
- expectation maximization
- multi frame
- image restoration
- bayesian classifiers
- prior knowledge
- em algorithm
- power law
- image denoising
- continuous data