Title :
A mixture-site model for edge-preserving image restoration
Author :
Fessler, Jeffrey A.
Author_Institution :
Michigan Univ., Ann Arbor, MI, USA
Abstract :
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
Keywords :
Bayes methods; edge detection; image restoration; medical image processing; positron emission tomography; single photon emission computed tomography; 3D data; Bayesian method; PET images; SPECT transmission images; binary mixture site process; edge-preserving image restoration; intensity process; line-site method; mixture-site model; noisy measurements; partial volume effects; region boundaries; Attenuation; Bayesian methods; Biological tissues; Biomedical imaging; Bones; Image restoration; Lungs; Pixel; Positron emission tomography; Single photon emission computed tomography;
Conference_Titel :
Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference
Conference_Location :
Austin, TX
Print_ISBN :
0-8186-6952-7
DOI :
10.1109/ICIP.1994.413867
