In this paper, we compare the performance of three iterative methods f
or image restoration: the Richardson-Lucy algorithm, the iterative con
strained Tikhonov-Miller algorithm (ICTM) and the Carrington algorithm
, Both the ICTM and the Carrington algorithm are based on an additive
Gaussian noise model, but differ in the way they incorporate the non-n
egativity constraint. Under low light-level conditions, this additive
(Gaussian) noise model is a poor description of the actual photon-limi
ted image recording, compared with that of a Poisson process. The Rich
ardson-Lucy algorithm is a maximum likelihood estimator for the intens
ity of a Poisson process, We have studied various methods for determin
ing the regularization parameter of the ICTM and the Carrington algori
thm and propose a (Gaussian) prefiltering to reduce the noise sensitiv
ity of the Richardson-Lucy algorithm. The results of these algorithms
are compared on spheres convolved with a point spread function and dis
torted by Poisson noise. Our simulations show that the Richardson-Lucy
algorithm, with Gaussian prefiltering, produces the best result in mo
st of the tests. Finally, we show an example of how restoration method
s can improve quantitative analysis: the total amount of fluorescence
inside a closed object is measured in the vicinity of another object b
efore and after restoration.