0

I have a scipy.spatial.Delaunay() triangulation, and I would like to remove 'skinny' triangles from it. How would I do this?

I have seen that some other software (like here: Houdini triangulate2d) have a 'minimum angle' option precisely to prevent skinny triangles. Is there some equivalent option for scipy.spatial.Delaunay(), or can someone provide a simple algorithm?

ap21
  • 2,372
  • 3
  • 16
  • 32
  • [Ruppert's algorithm](https://en.wikipedia.org/wiki/Ruppert%27s_algorithm) might be a good place to start. I think its the algorithm used in [MeshPy](https://pypi.org/project/MeshPy/) – DrBwts May 15 '19 at 15:19
  • @DrBwts Are `MeshPy` and `scipy.spatial.Delaunay()` interchangeable as far as Delaunay triangulations are concerned? – ap21 May 16 '19 at 15:18
  • not as far as I'm aware. MeshPy is basically a wrapper for [Triangle](http://www.cs.cmu.edu/~quake/triangle.html) (and also TetGen) which has much better control over the types of triangulation over domains with small angles. – DrBwts May 17 '19 at 10:42
  • @DrBwts Do you know of any other wrapper for `Triangle`? I could not figure out how to do a simple Delaunay triangulation of a point set using it. There's no obvious code for it. – ap21 May 17 '19 at 20:01
  • 1
    why not just use Triangle? – DrBwts May 18 '19 at 10:11
  • I just discovered that there is a simple wrapper for `Triangle` in python, also called `triangle`. – ap21 May 18 '19 at 15:55

0 Answers0