Fast Bayesian blind deconvolution with Huber Super Gaussian priors
- Beihang University, Beijing (China); ShanghaiTech University, Shanghai (China); Northwestern University
- University de Granada, Granada (Spain)
- Beihang University, Beijing (China)
- Northwestern University, Evanston, IL (United States)
We report expectation Maximization (EM) based inference has already proven to be a very powerful tool to solve blind image deconvolution (BID) problems. Unfortunately, three important problems still impede the application of EM in BID: the undesirable saddle points and local minima caused by highly nonconvex priors, the instability around zero of some of the most interesting sparsity promoting priors, and the intrinsic high computational cost of the corresponding BID algorithm. In this paper we first show how Super Gaussian priors can be made numerically tractable around zero by introducing the family of Huber Super Gaussian priors and then present a fast EM based blind deconvolution method formulated in the image space. In the proposed computational approach, image and kernel estimation are performed by using the Alternating Direction Method of Multipliers (ADMM), which allows to exploit the advantages of FFT computation. For highly nonconvex priors, we propose a Smooth ADMM (SADMM) approach to avoid poor BID estimates. In conclusion, extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art BID methods in terms of quality of the reconstructions and speed.
- Research Organization:
- Northwestern University, Evanston, IL (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- Grant/Contract Number:
- NA0002520
- OSTI ID:
- 1487466
- Alternate ID(s):
- OSTI ID: 1460440
- Journal Information:
- Digital Signal Processing, Journal Name: Digital Signal Processing Journal Issue: C Vol. 60; ISSN 1051-2004
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Variational Dirichlet Blur Kernel Estimation
A non-stationary image prior combination in super-resolution