mesh_surface_area

Calculates the surface area of a meshed region using scikit-image measure.mesh_surface_area method.

import matplotlib.pyplot as plt
import numpy as np
import porespy as ps
import inspect
inspect.signature(ps.metrics.mesh_surface_area)

<Signature (mesh=None, verts=None, faces=None)>

mesh

The input can be either mesh or verts and faces to calculate the surface area of a meshed region. mesh tuple can be generated using the mesh_region function. Let’s first create a mesh for a random blob:

np.random.seed(10)
im = ps.generators.blobs(shape=[50, 50, 50])
mesh = ps.tools.mesh_region(region=im)
surface_area = ps.metrics.mesh_surface_area(mesh=mesh)
print('surface area of void space is', str(surface_area))
fig, ax = plt.subplots(1, 1, figsize=[4, 4])
ax.imshow(im[:,:,5], origin='lower', interpolation='none')
ax.axis(False);

surface area of void space is 45683.796875

verts, faces

If mesh is not provided as the input, the verts and faces data must be given. verts and faces data of a meshed region can be calculated using porespy’s mesh_region function or directly from skimage.measure.marching_cubes function. Porespy’s mesh_region calls skimage.measure.marching_cubes method to mesh a region using marching cube algorithm.

verts, faces = mesh['verts'], mesh['faces']
surface_area = ps.metrics.mesh_surface_area(verts=verts, faces=faces)
print('surface area of void space is', str(surface_area))

surface area of void space is 45683.796875

Now let’s try this method on only a pore region of a segmented image for visualization purpose. Note that to calculate the surface area of each region in a segmented image, we can use porespy’s region_surface_areas method instead.

snow = ps.filters.snow_partitioning(im)
regions = snow.regions/im
region = regions==30
mesh = ps.tools.mesh_region(region=region)
ps.visualization.show_mesh(mesh)
area = ps.metrics.mesh_surface_area(verts=verts, faces=faces)
print('surface area of the meshed pore region '+str(area))

surface area of the meshed pore region 45683.796875