The paper summarizes a new Bayesian method for edge-preserving image restoration from noisy measurements. The line-site method of Geman and Geman (1984) forces region boundaries to lie along pixel boundaries, which is unnatural, particularly for 3D data. The present authors augment the intensity process with a binary "mixture site" process, which has one parameter for each pixel indicating the presence of a boundary at some unknown location within that pixel. The method was motivated by the PET and SPECT transmission images with partial volume effects, and is easily extended to 3D data sets.<>
A mixture-site model for edge-preserving image restoration
Proceedings of 1st International Conference on Image Processing ; 3 ; 162-166 vol.3
01.01.1994
558766 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
A Mixture-Site Model for Edge-Preserving Image Restoration
British Library Conference Proceedings | 1994
|Unsupervised edge-preserving image restoration via a saddle point approximation
British Library Online Contents | 1999
|Edge-preserved neural network model for image restoration
British Library Online Contents | 2001
|Image enhancement by edge-preserving filtering
IEEE | 1994
|Image Enhancement by Edge-preserving Filtering
British Library Conference Proceedings | 1994
|