Python scipy Delaunay cloud graphic display

Question

I have pointlist = [p1, p2, p3 ...] where p1 = [x1, y1], p2 = [x2, y2] ...

I want to use scipy.spatial.Delaunay to make a triangle on these point clouds, and then build it

How can i do this?

The documentation for Delaunay is really scarce

I still have this code

from subprocess import Popen, PIPE import os os.environ['point_num'] = "2000" cmd = 'rbox $point_num D2 | tail -n $point_num' sub_process = Popen(cmd, shell=True,stdout=PIPE,stderr=PIPE) output = sub_process.communicate() points = [line.split() for line in output[0].split('\n') if line] x = [p[0] for p in points if p] y = [p[1] for p in points if p] import matplotlib.pyplot as plt plt.plot(x,y,'bo') from scipy.spatial import Delaunay dl = Delaunay(points) convex = dl.convex_hull from numpy.core.numeric import reshape,shape convex = reshape(convex,(shape(convex)[0]*shape(convex)[1],1)) convex_x = [x[i] for i in convex] convex_y = [y[i] for i in convex] plt.plot(convex_x,convex_y,'r') plt.show() 

thanks

Answer 1

EDIT : also build a convex hull

 import numpy as np from scipy.spatial import Delaunay points = np.random.rand(30, 2) # 30 points in 2-d tri = Delaunay(points) # Make a list of line segments: # edge_points = [ ((x1_1, y1_1), (x2_1, y2_1)), # ((x1_2, y1_2), (x2_2, y2_2)), # ... ] edge_points = [] edges = set() def add_edge(i, j): """Add a line between the i-th and j-th points, if not in the list already""" if (i, j) in edges or (j, i) in edges: # already added return edges.add( (i, j) ) edge_points.append(points[ [i, j] ]) # loop over triangles: # ia, ib, ic = indices of corner points of the triangle for ia, ib, ic in tri.vertices: add_edge(ia, ib) add_edge(ib, ic) add_edge(ic, ia) # plot it: the LineCollection is just a (maybe) faster way to plot lots of # lines at once import matplotlib.pyplot as plt from matplotlib.collections import LineCollection lines = LineCollection(edge_points) plt.figure() plt.title('Delaunay triangulation') plt.gca().add_collection(lines) plt.plot(points[:,0], points[:,1], 'o', hold=1) plt.xlim(-1, 2) plt.ylim(-1, 2) # -- the same stuff for the convex hull edges = set() edge_points = [] for ia, ib in tri.convex_hull: add_edge(ia, ib) lines = LineCollection(edge_points) plt.figure() plt.title('Convex hull') plt.gca().add_collection(lines) plt.plot(points[:,0], points[:,1], 'o', hold=1) plt.xlim(-1, 2) plt.ylim(-1, 2) plt.show() 

Note that using scipy.spatial.Delaunay only to calculate a complex case is probably too large, because only the case can in principle be faster to calculate than triangulation. Unfortunately, Scipy does not have an interface, but for calculating cases directly with Qhull.