Two approaches for controlling noise amplification in image reconstruction.
Abstract
This paper presents two methods to overcome the problem of noise amplification during the image reconstruction process. The first method responds to the premise that the resolution sought in the reconstruction of different parts of an image should depend on the statistical quality of the corresponding data. The authors have developed a method that uses a Maximum a Posteriori Bayesian algorithm with Entropy prior (FMAPE), with hyperparameters that are region dependent. The second method is based on wavelet decomposition. After decomposing the raw image to be reconstructed into several wavelet planes and a residual image, the authors reconstruct each one independently using an iterativ Maximum Likelihood algorithm. The methods has been applied to real data from the initial (uncorrected) form of the Hubble Space Telescope.
- Publication:
-
Data Analysis in Astronomy
- Pub Date:
- 1997
- Bibcode:
- 1997daa..conf..311N
- Keywords:
-
- Image Reconstruction