Bayesian image processing of data from constrained source distributions-fuzzy pattern constraints

Publisher Website
Google Scholar
Abstract
A priori probability density functions characterising patterns which are imprecise spatially and with regard to amplitude (fuzzy patterns) and which are anticipated to be present in a radioisotopic source field were developed for use in Bayesian image processing (BIP). Corresponding iterative imaging algorithms were derived using the expectation maximisation (EM) technique of Dempster et al. BIP and standard non-BIP algorithms were applied to computer generated and experimental radioisotope phantom imaging data. Improved results were obtained with BIP.