def test_remove_mesh_boundary_faces(self):
mesh_a = self.cutSphere_A.copy()
mesh_a_save = mesh_a.copy()
boundary = stop.mesh_boundary(mesh_a)
mesh_processed = stop.remove_mesh_boundary_faces(mesh_a,
face_vertex_number=1)
# Non modification
assert (mesh_a.vertices == mesh_a_save.vertices).all()
# Correctness
# check that when removing all faces with any vertex on the boundary,
# all the boundary vertices are removed from the mesh
print(len(boundary[0]))
print(len(mesh_a.vertices))
print(len(mesh_processed.vertices))
assert (len(mesh_processed.vertices) == len(mesh_a.vertices) -
len(boundary[0]))
def generate_paraboloid_regular(A, nstep=50, ax=1, ay=1,
random_sampling=False,
random_distribution_type='gaussian',
ratio=0.1):
"""
generate a regular paraboloid mesh Z=K*Y^2
ratio and random_distribution_type parameters are unused if
random_sampling is set to False
:param A: amplitude of the paraboloid
:param nstep: nstepx or the sampling step stepx as a float !
:param ax: half length of the domain
:param ay: half width of the domain
:param random_sampling:
:param random_distribution_type:
:param ratio:
:return:
"""
# Parameters
xmin, xmax = [-ax, ax]
ymax = ay # ymin, ymax = [-ay, ay]
# Define the sampling
if isinstance(nstep, int):
stepx = (xmax - xmin) / nstep
else:
stepx = nstep
# Coordinates
x = np.arange(xmin, xmax, stepx)
# To generate y
y, curve_samples = adaptive_sampling(ymax, A, stepx)
X, Y = np.meshgrid(x, y)
X[::2] += stepx / 2
X = X.flatten()
Y = Y.flatten()
# Random perturbation
if random_sampling:
sigma = stepx * ratio # characteristic size of the mesh * ratio
nb_vert = len(x) * len(y)
if random_distribution_type == 'gamma':
theta = np.random.rand(nb_vert, ) * np.pi * 2
mean = sigma
variance = sigma ** 2
radius = \
np.random.gamma(mean ** 2 / variance, variance / mean, nb_vert)
X = X + radius * np.cos(theta)
Y = Y + radius * np.sin(theta)
elif random_distribution_type == 'uniform':
X = X + np.random.uniform(-1, 1, 100)
Y = Y + np.random.uniform(-1, 1, 100)
else:
X = X + sigma * np.random.randn(nb_vert, )
Y = Y + sigma * np.random.randn(nb_vert, )
# Delaunay triangulation: be careful, to do on the curvilinear aspects to
# avoid triangle flips
Xtmp, S = np.meshgrid(x, curve_samples)
S = S.flatten()
# faces_tri = Triangulation(X, S)
faces_tri = Delaunay(np.vstack((X, S)).T, qhull_options='QJ Qt Qbb')
# alternative setting? 'Qbb Qc Qz Qj'
Z = quadric(0, A)(X, Y)
coords = np.array([X, Y, Z]).transpose()
paraboloid_mesh = trimesh.Trimesh(faces=faces_tri.simplices,
vertices=coords,
process=False)
# remove the faces having any vertex on the boundary to avoid
# atypical faces geometry due to Delaunay triangulation in 2D
# TO DO: same boundary removal as for generate_quadric
return stop.remove_mesh_boundary_faces(paraboloid_mesh,
face_vertex_number=1)
visb_sc = splt.visbrain_plot(mesh=open_mesh, caption='open mesh')
# create points with vispy
for bound in open_mesh_boundary:
points = open_mesh.vertices[bound]
s_rad = SourceObj('rad', points, color='red', symbol='square',
radius_min=10)
visb_sc.add_to_subplot(s_rad)
lines = Line(pos=open_mesh.vertices[bound], width=10, color='b')
# wrap the vispy object using visbrain
l_obj = VispyObj('line', lines)
visb_sc.add_to_subplot(l_obj)
visb_sc.preview()
###############################################################################
# eroding the mesh by removing the faces having 3 vertices on the boundary
eroded_mesh = stop.remove_mesh_boundary_faces(open_mesh, face_vertex_number=1)
###############################################################################
# show the result
visb_sc2 = splt.visbrain_plot(mesh=eroded_mesh, caption='eroded mesh')
# show again the boundary of original mesh which have been removed with
# corresponding faces by the erosion
for bound in open_mesh_boundary:
points = open_mesh.vertices[bound]
s_rad = SourceObj('rad', points, color='red', symbol='square',
radius_min=10)
visb_sc2.add_to_subplot(s_rad)
lines = Line(pos=open_mesh.vertices[bound], width=10, color='b')
# wrap the vispy object using visbrain
l_obj = VispyObj('line', lines)
visb_sc2.add_to_subplot(l_obj)