Point cloud visualization

发布于 2022-08-08  691 次阅读


Please refresh the page if equations are not rendered correctly.
---------------------------------------------------------------

步骤:

  • 将点坐标存入ply文件;
  • 将向量数据以point_data的形式存入;
  • 使用ParaView的Glyph进行可视化即可。

实现代码:

import numpy as np
import meshio

vertices = np.random.rand(10, 3)

mesh = meshio.Mesh(points=vertices,
                   cells = [],
                   # Optionally provide extra data on points, cells, etc.
                   point_data={"nx": vertices[:,0], "ny": vertices[:,1], "nz": vertices[:,2]},
                   # Each item in cell data must match the cells array
                   # cell_data={"a": [[0.1, 0.2], [0.4]]},
                   )

meshio.write("test.ply", mesh, binary=False)

vtk

# vtk DataFile Version 4.0
vtk output
ASCII
DATASET POLYDATA
POINTS 20 double
0.00000000e+00  0.00000000e+00  0.00000000e+00 
3.30693964e-01  0.00000000e+00  1.62349735e-01 
6.61387927e-01  0.00000000e+00  3.07106356e-01 
9.92081891e-01  0.00000000e+00  4.18583239e-01 
1.32277585e+00  0.00000000e+00  4.84700133e-01 
1.65346982e+00  0.00000000e+00  4.98292247e-01 
1.98416378e+00  0.00000000e+00  4.57886663e-01 
2.31485774e+00  0.00000000e+00  3.67861955e-01 
2.64555171e+00  0.00000000e+00  2.37973697e-01 
2.97624567e+00  0.00000000e+00  8.22972951e-02 
3.30693964e+00  0.00000000e+00 -8.22972951e-02 
3.63763360e+00  0.00000000e+00 -2.37973697e-01 
3.96832756e+00  0.00000000e+00 -3.67861955e-01 
4.29902153e+00  0.00000000e+00 -4.57886663e-01 
4.62971549e+00  0.00000000e+00 -4.98292247e-01 
4.96040945e+00  0.00000000e+00 -4.84700133e-01 
5.29110342e+00  0.00000000e+00 -4.18583239e-01 
5.62179738e+00  0.00000000e+00 -3.07106356e-01 
5.95249134e+00  0.00000000e+00 -1.62349735e-01 
6.28318531e+00  0.00000000e+00 -1.22464680e-16 
LINES 1 21
20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

better than Ply format

import numpy as np
import meshio, re
from polykriging.utility import filenames
import pyvista as pv

path = ".04_coding/Python/00_Projects" \
       "/05_polyKriging/Data/processedData" \
       "/Vf57/Fabric/00_pointCloud/"
file_list = filenames(path, "ply")

# get the numerics in file_list using regular expression

yarn_index = np.array([re.findall(r'\d+', file_list[i])
                       for i in np.arange(len(file_list))], dtype=int)

for i, index in enumerate(yarn_index):

    mesh = meshio.read(path + file_list[i])

    try:
        vertices = np.vstack((vertices, mesh.points))
        labels = np.hstack((labels, index[0] * np.ones(mesh.points.shape[0])))

        print(index)
    except NameError:
        vertices = mesh.points
        labels = index * np.ones(mesh.points.shape[0])

    print(i)

labels = labels.astype(int)

# create a polydata object
point_cloud = pv.PolyData(vertices)
point_cloud.point_arrays['yarnIndex'] = labels

pv.set_plot_theme("paraview")
plotter = pv.Plotter()
plotter.add_mesh(point_cloud, point_size=3., render_points_as_spheres=True)
plotter.set_background("white")
# plotter.show_grid()
img = plotter.show(screenshot="point_cloud.png")

point_cloud.save("pc.vtk") # visualization in Paraview
Everything not saved will be lost.
最后更新于 2022-08-30