Ts. Zaccheo et Ra. Gonsalves, ITERATIVE MAXIMUM-LIKELIHOOD ESTIMATORS FOR POSITIVELY CONSTRAINED OBJECTS, Journal of the Optical Society of America. A, Optics, image science,and vision., 13(2), 1996, pp. 236-242
We present a unified approach for constructing iterative restorations
of positively constrained objects. Specifically, a set of nonlinear al
gorithms, one of which is the Richardson-Lucy algorithm, is described
for estimating positively constrained objects from data modeled by eit
her Poisson or Gaussian processes. Exponential and monomial functions
are used to remap the estimation space and to positively constrain the
restorations. This technique also provides a method to accelerate the
rate of convergence of known algorithms. Both one- and two-dimensiona
l examples are presented. (C) 1996 Optical Society of America