A half-quadratic block-coordinate descent method for spectral estimation

Signal Processing

Volumn 82, Issue 7, 2002, Pages 941-959

  Ciuciu, Philippe ^a   Idier, Jérôme ^b  

^a SERVICE HOSPITALIER FRÉDÉRIC JOLIOT   (France)

^b CNRS   (France)

Author keywords

Half quadratic regularization; Iteratively reweighted least squares; Legendre transform; Residual steepest descent; Spectral estimation

Indexed keywords

ALGORITHMS; ITERATIVE METHODS; LEAST SQUARES APPROXIMATIONS; MARKOV PROCESSES; SPECTRUM ANALYSIS;

SPECTRAL ESTIMATION;

SIGNAL PROCESSING;

EID: 0036644368     PISSN: 01651684     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0165-1684(02)00163-9     Document Type: Article

References (25)

1. A variational method in image recovery
2. Nondifferentiable optimization approximation
3. A generalized Gaussian image model for edge-preserving MAP estimation
4. Extrapolation and spectral estimation with iterative weighted norm modification
5. Two deterministic half-quadratic regularization algorithms for computed imaging
6. Deterministic edge-preserving regularization in computed imaging
7. Regularized estimation of mixed spectra using a circular Gibbs-Markov model
8. Globally convergent edge-preserving regularized reconstruction: An application to limited-angle tomography
9. Constrained restoration and the recovery of discontinuities
10. Nonlinear image recovery with half-quadratic regularization
11. Bayesian interpretation of periodograms
12. Analyse numèrique des inéquations variationnelles, tome 1: Théorie générale
13. Convex half-quadratic criteria and interacting auxiliary variables for image restoration
14. Spectrum analysis - A modern perspective
15. ∞ minimization via a variant of Karmarkar's algorithm
16. Interpolation and extrapolation using a high-resolution discrete Fourier transform
17. Iterative methods for total variation deionising
18. Fast, robust total variation-based reconstruction of noisy, blurred images
19. p deconvolution