flood

Floods each region with a specific value based on a specified statistical operation performed on values in that region.

Import packages

import numpy as np
import porespy as ps
import scipy.ndimage as spim
import matplotlib.pyplot as plt
import skimage
from edt import edt
ps.visualization.set_mpl_style()

[01:01:36] ERROR    PARDISO solver not installed, run `pip install pypardiso`. Otherwise,          _workspace.py:56
                    simulations will be slow. Apple M chips not supported.                                         

im

The distance transform can have statistical calculations performed

im = ps.generators.blobs(shape=[200, 200])
dt = edt(im)

plt.figure(figsize=[6, 6])
plt.imshow(dt/im)
plt.axis(False);

labels

snow_partitioning can be used to create regions

regions = ps.filters.snow_partitioning(im, r_max=4, sigma=0.4)
labels = regions.regions

plt.figure(figsize=[6, 6])
plt.imshow(labels/im)
plt.axis(False);

mode

Various functions in scipy.ndimage.measurements are called to perform statistical calculation. The mode indicates which function to call.

x1 = ps.filters.flood(im=dt, labels=labels, mode='max')
x2 = ps.filters.flood(im=dt, labels=labels, mode='mean')
x3 = ps.filters.flood(im=dt, labels=labels, mode='sum')

fig, ax = plt.subplots(1, 3, figsize=[18, 18])
ax[0].imshow(x1)
ax[0].axis(False)
ax[0].set_title('mode = max')
ax[1].imshow(x2)
ax[1].axis(False)
ax[1].set_title('mode = mean')
ax[2].imshow(x3)
ax[2].axis(False)
ax[2].set_title('mode = sum');