This routine converts the RESULTS output from PISAFIND into an ARD
description. The ARD description consists of a series of ellipses,
one for each detected object, which can be used to identify the
area of the images on a frame.
The scale of the ellipses can be changed allowing the effective
areas to be modified.