<div dir="ltr">Philippe-<div><br></div><div style>At some point can you send me the code / data files / whatever you use to generate some of these tests (particularly the star). I'll be using them to test the Delaunay code of course. Thanks,</div>
<div style><br></div><div style>W</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sat, Mar 9, 2013 at 7:18 AM, Philippe Pébay <span dir="ltr"><<a href="mailto:philippe.pebay@kitware.com" target="_blank">philippe.pebay@kitware.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hello all,<br><br>I think that I just ran into a numerical instability of the Delanauy 2D triangulator, when used with a specified boundary constraint. For those interested in this issue, please look at the attached picture: the boundary constraint is in the upper left corner, which also includes the generators in red. The CDT is shown in the upper right corner, where I indicated the problematic edge: clearly, the 2 adjacent triangles do not satisfy the empty ball property, and that edge should be flipped. This is confirmed by the Voronoi diagram underneath where we see one cell protruding into another. <br>
<br>In fact the highlighted edge should be flipped to obtain a correct CDT, and this what happens indeed when I nudge on of the points with a tiny epsilon. See the correct picture attached (where the "forced edges" for the sake of the constraint have been removed).<br>
<br>By the way I observed similar instabilities, but with <u>slightly different generators</u>, when calling the Delaunay 2D from ParaView.<br><br>I intend to work on the Delaunay 2D triangulator, but in collaboration with whoever is interested in this. Please let me know.<br>
<br>Thanks<span class="HOEnZb"><font color="#888888"><br>P<br clear="all"><br>-- <br><font color="#888888">Philippe Pébay, PhD<br></font><font color="#888888">Director of Visualization and High Performance Computing /<br>
</font><font color="#888888">Directeur de la Visualisation et du Calcul Haute Performance<br>
Kitware SAS<br>26 rue Louis Guérin, 69100 Villeurbanne, France</font><br>
<font color="#888888"><a value="+33426685003">+33 (0) 6.83.61.55.70 / 4.37.45.04.15</a></font><font color="#888888"><br><a href="http://www.kitware.fr/" target="_blank">http://www.kitware.fr</a></font><font color="#888888"><a href="http://www.kitware.fr/" target="_blank"></a></font>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br>William J. Schroeder, PhD<br>Kitware, Inc.<br>28 Corporate Drive<br>Clifton Park, NY 12065<br><a href="mailto:will.schroeder@kitware.com">will.schroeder@kitware.com</a><br>
<a href="http://www.kitware.com">http://www.kitware.com</a><br>(518) 881-4902
</div>