[Octave-bug-tracker] [bug #33923] problem with voronoin output

[Yulin Chang]

URL:
  <http://savannah.gnu.org/bugs/?33923>

                 Summary: problem with voronoin output
                 Project: GNU Octave
            Submitted by: ychang
            Submitted on: Tue 02 Aug 2011 02:45:06 AM GMT
                Category: None
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Incorrect Result
                  Status: None
             Assigned to: None
         Originator Name: Yulin Chang
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 3.2.3
        Operating System: GNU/Linux

    _______________________________________________________

Details:

he output of the indices (F) of the voronoi diagrams is messed up:
[C,F] = voronoin(xy);

Specifically, an element of F should be the indices of those vertices
in C that form a facet (or cell called by matlab) surrounding the
correponding point in xy, i.e., the elements in F and the input points
in xy should have a one-to-one correspondence in the same order
(first element of F corresponds to first element of xy, second element
of F corresponds to second element of xy, etc.). The order is messed up,
and I can't for the life of me figure out how the elements in F are
organized. Since the main purpose of voronoin is to compute the area
surrounding each point in xy, the incorrect order makes the function
almost useless.

The problem can be checked using the following code.


x=[6,9,1,8,10,3,5,2,7,4];
y=[4,6,8,9,1,3,2,10,7,5];
[cx,cy]=voronoi(x,y);
figure; plot(x,y,'ro',cx,cy,'b-'); axis([0 11 0 11]);
% you can see the first point (6,4) is surrounded by a pentagon
xy=[x(:),y(:)];
[C,F]=voronoin(xy);
% check F, the firs element contains only three indices, clearly not
% for the first element of xy, which is (6,4). And the correct element
% for (6,4) should, obviously, contain 5 indices






    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?33923>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/