BDD algorithm for image space of a finite function

I asked this already in comp.theory but did not get an answer

For a current project I needed a way to determine the
size of the image space of a finite function and came up with the
following algorithm:

I use BDDs  (more precisely Reduced Ordered Binary Decision Diagrams)
to represent the function f.
Hence, we have a vector F=(f_0,...,f_{k-1}) which represents the
function.
The enumeration is a recursive process:

enum_space(F) = enum_space_recursive(f,0,true);

enum_space_recursive(f,i,cond)
BEGIN

 if (i>= k) return 1
 else

res_true = 0; res_false=0;
 cond_true = cond AND f_i;
    if (cond_true <> FALSE)
       res_true = enum_space_recursive(f,i+1,cond_true);
    cond_false = cond AND  (NOT f_i);
    if (cond_false <> FALSE)
       res_true = enum_space_recursive(f,i+1,cond_false);

return res_true + res_false;
END;

Note that cond, cond_true, and cond_false are functions as well
and that the cmparisons with FALSE represent a satisfiability test.
For the cases I needed the method it works quite well and is
much faster than enumerating the image space
by brute force enumeration of the function domain and evaluation of
the function.
I have used the BDD-package CUDD.

Is this method known and published?
Maybe it has even a partuclar name?
I would like to give a reference even  if I came up with it
independently.

Thanks in advance,
Andreas