[vtkusers] weird patches in delaunay triangulation

[John Biddiscombe]

>From a quick look at the images, it looks like you've got a possible 
case of duplicated poiunts (which you have removed), but with different 
scalar values attached to the originals. when you remove the duplicated 
points, you may be removing a doof scalar value rather than a bad one 
(?). Can you remove points and average the scalars that exist at that 
point.

Perhaps I'm guessing badly, but it looks suspicious that the meshes 
(wireframe in images 2,3) have slightly different colours around the 
anomalous regions which makes it looks like selective scalar removal.

JB


Burlen wrote:
> Hi, in an attempt to visualize some multi resolution data I'm 
> constructing a surface by passing vtkPolyData made from vtkPoints, & a 
> vtkDoubleArray through the vtkDelaunay2D filter & then writing the 
> output to file for use with paraview. In the resulting visualization I 
> see a bunch of anomalies, that look like pentagonal patches in one 
> case and hexagonal in another. the following images shows what I'm 
> talking about. There are three visualizations shown, the first is that 
> of a vtkUnstructuredGrid obtained from the output of 
> vtkAppendFilter(), the second is that of a vtkDelaunay2D filter acting 
> on the vtkPoints and arrays from vtkPointData that are in the output 
> of the vtkAppendFilter, the third is that of a vtkDelaunay2D acting on 
> a vtkPolyData which I built using vtkPoints and vtkDoubleArrays sorted 
> in ascending order after removing duplicate points.
> http://quaoar.sr.unh.edu/images/delaunay_splotches.png
> Does any one know what I could do to correct this? Or why it might be 
> happening?
>
>   


-- 
John Biddiscombe,                            email:biddisco @ cscs.ch
http://www.cscs.ch/about/BJohn.php
CSCS, Swiss National Supercomputing Centre  | Tel:  +41 (91) 610.82.07
Via Cantonale, 6928 Manno, Switzerland      | Fax:  +41 (91) 610.82.82