Computing Surface Normals

Compute normals on a surface.

import numpy as np

from pyvista import examples

mesh = examples.download_topo_global()
mesh.plot(cmap="gist_earth", show_scalar_bar=False)

Now we have a surface dataset of the globe loaded - unfortunately, the dataset shows the globe with a uniform radius which hides topographic relief. Using pyvista.PolyData.compute_normals(), we can compute the normal vectors on the globe at all points in the dataset, then use the values given in the dataset to warp the surface in the normals direction to create some exaggerated topographic relief.

# Compute the normals in-place and use them to warp the globe
mesh.compute_normals(inplace=True)  # this activates the normals as well

Header	Data Arrays
PolyData Information N Cells 2333880 N Points 2336041 N Strips 0 X Bounds -1.000e+00, 1.000e+00 Y Bounds -1.000e+00, 1.000e+00 Z Bounds -1.000e+00, 1.000e+00 N Arrays 3	Name Field Type N Comp Min Max altitude Points float32 1 -1.042e+04 6.527e+03 Normals Points float32 3 -1.000e+00 1.000e+00 Normals Cells float32 3 -1.000e+00 1.000e+00

Now use those normals to warp the surface

warp = mesh.warp_by_scalar(factor=0.5e-5)

And let’s see it!

warp.plot(cmap="gist_earth", show_scalar_bar=False)

We could also use face or cell normals to extract all the faces of a mesh facing a general direction. In the following snippet, we take a mesh, compute the normals along its cell faces, and extract the faces that face upward.

mesh = examples.download_nefertiti()
# Compute normals
mesh.compute_normals(cell_normals=True, point_normals=False, inplace=True)

# Get list of cell IDs that meet condition
ids = np.arange(mesh.n_cells)[mesh['Normals'][:, 2] > 0.0]

# Extract those cells
top = mesh.extract_cells(ids)

cpos = [
    (-834.3184529757553, -918.4677714398535, 236.5468795300025),
    (11.03829376004883, -13.642289291587957, -35.91218884207208),
    (0.19212361465657216, 0.11401076390090074, 0.9747256344254143),
]

top.plot(cpos=cpos, color=True)