10
Does anyone know of a Python replacement for Matlab / Octave bwdist() function? This function returns Euclidian distance of each cell to the closest non-zero cell for a given matrix. I saw an Octave C implementation, a pure Matlab implementation, and I was wondering if anyone had to implement this in ANSI C (which doesn't include any Matlab / Octave headers, so I can integrate from Python easily) or in pure Python.
Both links I mentioned are below:
As a test, a Matlab code / output looks something like this:
bw= [0 1 0 0 0;
1 0 0 0 0;
0 0 0 0 1;
0 0 0 0 0;
0 0 1 0 0]
D = bwdist(bw)
D =
1.00000 0.00000 1.00000 2.00000 2.00000
0.00000 1.00000 1.41421 1.41421 1.00000
1.00000 1.41421 2.00000 1.00000 0.00000
2.00000 1.41421 1.00000 1.41421 1.00000
2.00000 1.00000 0.00000 1.00000 2.00000
I tested a recommended distance_transform_edt call in Python, which gave this result:
import numpy as np from scipy import ndimage
a = np.array(([0,1,0,0,0],
[1,0,0,0,0],
[0,0,0,0,1],
[0,0,0,0,0],
[0,0,1,0,0]))
res = ndimage.distance_transform_edt(a)
print res
[[ 0. 1. 0. 0. 0.]
[ 1. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 1.]
[ 0. 0. 0. 0. 0.]
[ 0. 0. 1. 0. 0.]]
This result does not seem to match the Octave / Matlab output.