Line segments not coloured correctly with data read in from VTU files

Hi @marcomusy,

Another VTU-related problem. I was trying to plot line segments with different colors used to denote different lithology units as revealed from drilling. The original data was prepared into vtu files with segments as cells. The following picture was originally plotted using PyVista:

Not something I am satisfied with but at least it is able to show segments with different colors. I then tried to plot something similar using vedo and here’s what I got:

As you can see, the segments are not properly colored as expected although the colorbar was showing correctly. Any ideas on how to get the expected picture plotted? Thanks very much!

Here are all the data needed to plot the picture: vedo_issue.zip

The following code was used to plot the figure:

from vedo import *
import numpy as np

settings.defaultFont = 'Theemim'
settings.multiSamples=8

# Plot the two drillholes
l2400_vtu = 'L2400E.vtu'
l3200_vtu = 'L3200E.vtu'

l2400 = UGrid(l2400_vtu).tomesh().lineWidth(20).lc('w')
l3200 = UGrid(l3200_vtu).tomesh().lineWidth(20)

# Lookup table for drillhole lithology
lut_table = [
    # Value, color, alpha, category
    (1., 'gray', 1, 'Overburden'),
    (2., 'bronw', 1, 'Fault'),
    (3., 'red', 1, 'Fault (graphite)'),
    (4., 'green', 1, 'Basement'),
]
lut = buildLUT(lut_table)
l2400.cmap(lut, 'Lithology', on='cells')
l3200.cmap(lut, 'Lithology', on='cells')
l2400.addScalarBar3D(
    categories=lut_table,
    title="Lithology",
    labelSize=2,
    pos=(606500, 6371000, -50),
    sx=40,
    sy=1200,
)
# l3200.addScalarBar3D(
#     categories=lut_table,
#     title="Lithology",
#     labelSize=2,
#     pos=(606500, 6371500, -50),
# )

# Create a UGrid object from surface_input.vtu file
igen = 20
vtu_file = 'test_knots_igen=' + '{:03d}'.format(igen) + '.vtu'
ug_cell = UGrid(vtu_file)
ug_node = ug_cell.clone()

# Get the mesh objects for both cells and nodes
mesh_cell = ug_cell.tomesh().lineWidth(5).lineColor('blue').alpha(0.5)
mesh_cell.cmap("coolwarm", 'names001', on='cells')
mesh_cell.addScalarBar3D(
    title='Cell conductivity (S/m)',
    titleXOffset=-2.5,
    labelSize=1.5,
    pos=(603550, 6371500, -50),
    sx=30,
    sy=1200,
)

mesh_node = ug_node.tomesh()

# We need to build a look up table for our color bar, and now it supports
# using category names as labels instead of the numerical values
# This was implemented upon my request
lut_table = [
    # Value, color, alpha, category
    (1.2, 'cyan', 1, r'\pm200'),
    (2.2, 'red', 1, '\pm200'),
    (3.2, 'dodgerblue', 1, '\pm400'),
    (4.2, 'purple', 1, '\pm500'),
    (5.2, 'green', 1, '\pm150'),
]
lut = buildLUT(lut_table)

# Adde color maps to both nodes and cells
mesh_node.cmap(lut, 'nodeGroup', on='points')
mesh_node.pointSize(25)

# Options for axes
group = mesh_node + mesh_cell
axes = Axes(group,
            xtitle='Easting (m)',
            ytitle='Northing (m)',
            ztitle='Elevation (m)',
            xTitlePosition=0.55,
            xLabelSize=0.018,
            yTitlePosition=0.65,
            yLabelRotation=-90,
            yLabelSize=0.018,
            # yLabelOffset=-1.5,
            # yTitleOffset=-1.12,
            yTitleRotation=180,
            axesLineWidth=4,
            xrange=mesh_cell.xbounds(),
            yrange=mesh_cell.ybounds(),
            # yzShift=1,
            zxGrid=True,
            )
# manually move the Z axis back in place
for a in axes.unpack():
    # print(a.name)
    if ("zAxis" in a.name or "zM" in a.name or "zN" in a.name
        or "zxGrid" in a.name or "zTipCone" in a.name or 'ztitle' in a.name):
        yb = group.ybounds()
        a.shift(0, (yb[1]-yb[0]), 0)

mesh_node.addScalarBar3D(
    categories=lut_table,
    title='Search volumes (m)',
    titleRotation=180,
    titleXOffset=-4,
    labelSize=2,
    labelOffset=1.5,
    labelRotation=-90,
    pos=(606000, 6371000, -50),
    sx=40,
    sy=1200,
)
# put scalarbar vertical and shift it up a bit
mesh_node.scalarbar.rotateZ(90, locally=True).shift(-1200, -1000, 0)
# mesh_node.scalarbar.shift(-1000, -500, 0)

# Also rotate the scalarbar showing cell conductivity
mesh_cell.scalarbar.rotateX(90, locally=True).rotateY(-45, locally=True).rotateZ(-15, locally=True)

plt = Plotter()
plt.camera.SetPosition( [602592.516, 6368179.979, 2912.739] )
plt.camera.SetFocalPoint( [604701.971, 6370768.195, -30.009] )
plt.camera.SetViewUp( [0.472, 0.466, 0.748] )
plt.camera.SetDistance( 4450.667 )
plt.camera.SetClippingRange( [2453.388, 6975.564] )
# Define the window size in pixels
# size = [3940, 2160]
size = [2560, 1600]
# plt.show(mesh_node, mesh_cell, axes, l2400, l3200, size=size, zoom=1.2,
#          interactive=False)
plt.show(mesh_node, mesh_cell, axes, l2400, l3200, size=size,
         interactive=True)

# filename = 'surface_input_no_point_size.png'
filename = 'surface_input.png'
screenshot(filename)

from wand import image

with image.Image(filename=filename) as imag:
    imag.trim(color=None, fuzz=0)
    imag.save(filename=filename)

[marcomusy]

Cool,

The reason why some of the edges disappeared was because of VTK? As long as there is a solution to it, I can keep using it.

Yes. Sadly VTK starting from version 8.2 suffers of artifacts when plotting lines and transparent objects. Shrinking a bit the triangles is probably helping with that…

The program for the above rendering is this:

#!/usr/bin/env python3
#
from vedo import *

settings.defaultFont = 'Theemim'
settings.multiSamples= 8
wsize = [2060, 1200]

# Plot the two drillholes
l2400_vtu = 'L2400E.vtu'
l3200_vtu = 'L3200E.vtu'

# Lookup table for drillhole lithology
lut_table = [
    # Value, color, alpha, category
    (1., 'gray',  1, 'Overburden'),
    (2., 'bronw', 1, 'Fault'),
    (3., 'red',   1, 'Fault\n(graphite)'),
    (4., 'green', 1, 'Basement'),
]
lut = buildLUT(lut_table)

l2400 = UGrid(l2400_vtu).tomesh().clean().lineWidth(20)
l2400.cmap(lut, 'Lithology', on='cells')
l2400.property.RenderLinesAsTubesOn()
l2400.addScalarBar3D(
    categories=lut_table,
    title="Lithology",
    titleSize=2,
    titleRotation=180,
    titleXOffset=-2.,
    labelSize=2,
    pos=(606300, 6371000, -50),
    sx=60,
    sy=800,
)
l2400.scalarbar.rotateZ(-45, locally=True).rotateX(45, locally=True)

l3200 = UGrid(l3200_vtu).tomesh().clean().lineWidth(20)
l3200.cmap(lut, 'Lithology', on='cells')
l3200.property.RenderLinesAsTubesOn()
# no need of scalarbar here

# Create a UGrid object from surface_input.vtu file
igen = 20
vtu_file = 'test_knots_igen=' + '{:03d}'.format(igen) + '.vtu'
ug_cell = UGrid(vtu_file)
ug_node = ug_cell.clone()

# Get the mesh objects for both cells and nodes
mesh_cell = ug_cell.tomesh(shrink=0.85).clean().alpha(0.6) #see below
mesh_cell.cmap("coolwarm", 'names001', on='cells').lighting("off")
mesh_cell.addScalarBar3D(
    title='Cell conductivity (S/m)',
    titleSize=2,
    titleRotation=180,
    titleXOffset=-3,
    labelSize=1.5,
    pos=(603550, 6371000, 0),
    sx=40,
    sy=1000,
)

# overlay the lines manually because vtk doesn't do a good job with anitialiasing
mesh_wire = Mesh(ug_cell.tomesh().clean()).wireframe()
mesh_wire.lineWidth(3).lineColor('blue4').lighting("off")

# We need to build a look up table for our color bar, and now it supports
# using category names as labels instead of the numerical values
lut_table = [
    # Value, color, alpha, category
    (1.2, 'cyan', 1, r'\pm200'),
    (2.2, 'red', 1, '\pm200'),
    (3.2, 'dodgerblue', 1, '\pm400'),
    (4.2, 'purple', 1, '\pm500'),
    (5.2, 'green', 1, '\pm150'),
]
lut = buildLUT(lut_table)

mesh_node = ug_node.tomesh().clean()

# Add color maps to both nodes and cells
mesh_node.cmap(lut, 'nodeGroup', on='points').pointSize(25)
mesh_node.addScalarBar3D(
    categories=lut_table,
    title='Search volumes (m)',
    titleRotation=180,
    titleXOffset=-4,
    labelSize=2,
    labelOffset=1.5,
    labelRotation=-90,
    pos=(606000, 6371000, -50),
    sx=40,
    sy=1200,
)
# put scalarbar vertical and shift it up a bit
mesh_node.scalarbar.rotateZ(90, locally=True).shift(-1200, -1000, 0)

# Options for axes
group = mesh_node + mesh_cell
axes = Axes(group,
            xtitle='Easting (m)',
            ytitle='Northing (m)',
            ztitle='Elevation (m)',
            xTitlePosition=0.55,
            xLabelSize=0.018,
            yTitlePosition=0.65,
            yLabelRotation=-90,
            yLabelSize=0.018,
            yTitleRotation=180,
            zAxisRotation=-40,
            axesLineWidth=4,
            xrange=mesh_cell.xbounds(),
            yrange=mesh_cell.ybounds(),
            zShiftAlongY=1.0,
            zxGrid2=True,
            xyFrameLine=True,
            gridLineWidth=2,
)

plt = Plotter(size=wsize)
plt.camera.SetPosition( [602867.008, 6368629.132, 2610.594] )
plt.camera.SetFocalPoint( [604678.698, 6370852.003, 83.235] )
plt.camera.SetViewUp( [0.446, 0.489, 0.75] )
plt.camera.SetDistance( 3822.423 )
plt.show(mesh_node, mesh_cell, mesh_wire, l2400, l3200, axes, resetcam=False)

[XushanLu]

Ah, yes! That should be a cmap thing. Tried it and it surely works. Thanks!