A. Pizurica: research interests
Multiresolution Statistical Image Models
I am interested in statistical modeling of images and their characteristic features at multiple resolution scales.
This includes statistical characterization of image discontinuities and image noise in the wavelet and other related x-let representations.
My results in this area are mostly related to
- Statistical characterization of a signal of interest in the presence of noise.
- Bernoulli Laplacian i generalized Laplacian priors for image wavelet (and x-let) coefficients
- Empirical Bayesian models for image wavelet coefficients
- Statistical models for spatial clustering of image wavelet coefficients
- Local regularity estimators in the wavelet domain and their statistical characterization
- Statistical models for spatial clustering of image wavelet (and x-let) coefficients
- Relations with robust statistics and deriving edge stopping functions for nonlinear diffusion
Selected related publications
-
A. Pizurica, I. Vanhamel, H. Sahli, and A. Philips, W.and Katartzis,
"A Bayesian approach to nonlinear diffusion based on a Laplacian prior for
ideal image gradient",
in IEEE Workshop on Statistical Signal Processing SSP 2005,
Bordeaux, France, July 17-20, 2005.
-
A. Pizurica, I. Vanhamel, H. Sahli, W. Philips, and A. Katartzis,
"A Bayesian formulation of edge-stopping functions in nonlinear diffusion",
IEEE Signal Processing Letters, vol. 8, no. 13, pp. 501-504, 2006.