Vol. 10, Issue 4, pp. 351-372

Abstract

The numerical properties of the direct recovery method are discussed from the viewpoint of their importance in the problem of the image intensity reconstruction from the discrete and finite set of measurement results obtained by a square- law sampling system. The effect of many factors like: number of sampling points, value of sampling step, sizes of the integrating element and the like, on the numerical aspects of the procedure considered is examined. Both one- and two-dimensional cases are analyzed. It has been shown that the direct recovery method is very sensitive to even very small changes in the intensity spread function.