Filling a specific shape within a numpy array

Question

I have a 2d numpy array and would like to increase the values of a specific part of it by 1. A simple example is:

Empty array:

[[0 0 0 0 0]
 [0 0 0 0 0]
 [0 0 0 0 0]
 [0 0 0 0 0]
 [0 0 0 0 0]]

We decide to base our area on e.g. 3 points ( [[0,2], [2,4], [3,1]] ) but the number of points could be higher as well:

[[0 0 1 0 0]
 [0 0 0 0 0]
 [0 0 0 0 1]
 [0 1 0 0 0]
 [0 0 0 0 0]]

And fill the shape within this polygone with ones:

[[0 0 1 0 0]
 [0 0 1 1 0]
 [0 1 1 1 1]
 [0 1 1 0 0]
 [0 0 0 0 0]]

How can I do this in an automated way? Something like:

arr = np.zeros([5,5])
shapePoints = [[0,2], [2,4], [3,1]]
valueToFill = 1
arr = fillShape(arr, shapePoints, valueToFill)

... where fillShape is the function that I am searching for. That must exist in numpy?

Whether the number of vertices you are trying to fill is fixed (Vertices:Three, Shape:Triangle in this case) or can it change in the future? — codeslord
– codeslord, Commented Apr 23, 2019 at 12:44
Yes, it can change - minimum of 3 and maximum of possibly 20 or so — lordy
– lordy, Commented Apr 23, 2019 at 12:46
Have you tried Cairo, I do not know exactly why you want this, but cairo allows creating surfaces. cairographics.org — codeslord
– codeslord, Commented Apr 23, 2019 at 12:52
It is in the end not so much about the plotting but about creating the values of the np array ... I can plot all my shapes using PolygonPatch but I want to map them onto an np array in order to do further calculations on them — lordy
– lordy, Commented Apr 23, 2019 at 12:57

Mihai Andrei · Accepted Answer · 2019-04-23 16:20:47Z

2

Scikit has a implementation for such a polygon drawing algorithm. See skimage.draw.polygon

Edit: concrete code for the question

from skimage.draw import polygon, polygon_perimeter
import numpy as np
arr = np.zeros((5, 5))
shapePoints = np.array([[0,2], [2,4], [3,1]])
points_r, points_c = shapePoints[:, 0], shapePoints[:, 1]
interior_r, interior_c = polygon(points_r, points_c)
perimeter_r, perimeter_c = polygon_perimeter(points_r, points_c)
arr[perimeter_r, perimeter_c] = 3
arr[points_r, points_c] = 2
arr[interior_r, interior_c] = 1

Note that this produces a different result than the one in the question in order to illustrate the difference between interior and perimeter.

>>> arr
array([[0., 0., 2., 0., 0.],
       [0., 0., 1., 3., 0.],
       [0., 3., 1., 1., 2.],
       [0., 2., 3., 0., 0.],
       [0., 0., 0., 0., 0.]])

edited Apr 23, 2019 at 16:20

answered Apr 23, 2019 at 13:08

Mihai Andrei

1,0448 silver badges11 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

kmario23 Over a year ago

please consider adding the solution here, for the benefit of everyone

Mihai Andrei Over a year ago

That would be just copy-pasting the documentation example. Is that encouraged here?

kmario23 Over a year ago

better than not having anything, IMO :)

Collectives™ on Stack Overflow

Filling a specific shape within a numpy array

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related