Python: Get all coordinates between two 3D points or draw a 3D Line

Question

I have the following problem. I have 3D point coordinates and I want to conenct them in an array or "draw a line" like you can do it in 2D with skimage(http://scikit-image.org/docs/0.13.x/api/skimage.draw.html#skimage.draw.line). The optimal case would be if i could directly draw a cylinder with a radius in an array and give different radius different values like rings arround the line.(wrinkles could be a problem here). There have been approaches to do that but they are not the right thing i think like here:

"Draw" a 3d line into an array

With the two approaches of @Paul Panzer (Fastest way to get all the points between two (X,Y) coordinates in python) you get all coordinates between two 2D Points, but how would it look like in 3D, esspecially the second approach which is faster?:

import numpy as np
from timeit import timeit

def connect(ends):
    d0, d1 = np.abs(np.diff(ends, axis=0))[0]
    if d0 > d1: 
        return np.c_[np.linspace(ends[0, 0], ends[1, 0], d0+1, dtype=np.int32),
                     np.linspace(ends[0, 1]+0.5, ends[1, 1]+0.5, d0+1, dtype=np.int32)]
    else:
        return np.c_[np.linspace(ends[0, 0]+0.5, ends[1, 0]+0.5, d1+1, dtype=np.int32),
                     np.linspace(ends[0, 1], ends[1, 1], d1+1, dtype=np.int32)]


def connect2(ends):
    d0, d1 = np.diff(ends, axis=0)[0]
    if np.abs(d0) > np.abs(d1): 
        return np.c_[np.arange(ends[0, 0], ends[1,0] + np.sign(d0), np.sign(d0), dtype=np.int32),
                     np.arange(ends[0, 1] * np.abs(d0) + np.abs(d0)//2,
                               ends[0, 1] * np.abs(d0) + np.abs(d0)//2 + (np.abs(d0)+1) * d1, d1, dtype=np.int32) // np.abs(d0)]
    else:
        return np.c_[np.arange(ends[0, 0] * np.abs(d1) + np.abs(d1)//2,
                               ends[0, 0] * np.abs(d1) + np.abs(d1)//2 + (np.abs(d1)+1) * d0, d0, dtype=np.int32) // np.abs(d1),
                     np.arange(ends[0, 1], ends[1,1] + np.sign(d1), np.sign(d1), dtype=np.int32)]


ends = np.array([[ 1520, -1140],
                 [ 1412,  -973]])

score 0 · Answer 1 · answered Feb 10 '18 at 09:03

0

Since you tagged the question 'vtk', I assume you can use it in your code. In that case, I think vtkTubeFilter does exactly what you are looking for - you create a vtkPolyData with all your line segments (you have your endpoints for that) and for each segment, the filter generates a cylinder. You can also specify different radius for each segment as you wanted: call SetVaryRadiusToVaryRadiusByScalar() on the filter to turn it on and provide the array with radii as scalar data.

You can find a code example that does all that (and more) here: https://www.vtk.org/Wiki/VTK/Examples/Cxx/VisualizationAlgorithms/TubesWithVaryingRadiusAndColors

answered Feb 10 '18 at 09:03

tomj

1,089
7
17

I used the TubeFilter before, but I think the problem here is that theTubeFilter only is the contour of the cylinder. To get all coordinates in the inner of the Tube I have to use vtk.vtkSelectEnclosedPoints() which is very time consuming. – Liwellyen Feb 10 '18 at 12:41
I see, I misunderstood the problem I think, you are working with discrete volumetric data and you want to fill all voxels along the line(s)...in that case, I can't help :( – tomj Feb 10 '18 at 19:00

Python: Get all coordinates between two 3D points or draw a 3D Line

1 Answers1