I am trying to build a Delaunay triangulation for a very specific case where the input coordinates x and y are orthogonal and relatively equidistant.
Given that the data size is relatively large (the triangulation points are 1000x1200) and that the Qhull algorithm is not aware of my additional orthogonal condition, the triangulation is relatively slow (25 seconds on my machine).
Thus, I would like to manually construct a Delaunay triangulation with each of my famous ATVs divided into two triangles. I admit that this will not always lead to actual Delaunay triangulation (for example, when the steps x and y are significantly different), but in my case I'm pretty sure that the sub-structural approach will lead to good triangulation.
In the following graph, I designated each of the triangles with an index, the initial direction for determining the vertices and vertices:
In this case, I have both x and y coordinates [-1, 1.33, 3.67, 6]and, [2, 4.5, 7, 9.5, 12]respectively.
SciPy Qhull, , equations ( http://docs.scipy.org/doc/scipy-dev/reference/generated/scipy.spatial.ConvexHull.html).
, , , , paraboloid_scale paraboloid_shift, , Qhull. n_dimensions + 1 , SciPy , :
dist = d.equations[isimplex*(d.ndim+2) + d.ndim+1]
for k in xrange(d.ndim+1):
dist += d.equations[isimplex*(d.ndim+2) + k] * point[k]
, :
equation?- , ?
equation, , Qhull?