Abstract
Nuclear medicine imaging systems rely on photon detection as the basis of image formation. One of the major sources of error in these imaging systems is Poisson noise. In this paper, we develop a novel multiscale image restoration procedure for photon imaging systems. It consists in separating in the wavelet-domain data points which belong to structures from those due to the noisy background. The latter are replaced by coefficients obtained by the introduction of a multiresolution regularity constraint. The restored image represents therefore a compromise between fidelity of the data and fidelity of the a priori knowledge on the smoothness of the solution. The performance of the method is assessed with simulated data experiments.

This publication has 8 references indexed in Scilit: