How to add a colorbar to GLViewWidget()

[Igor Shovkun]

Hi everybody!

I am working on visualization of 3D data.  I have a GLViewWidget widget, and I added a GLMeshItem to plot the data computed on a mesh.

I have managed to plot and color it using pg.ColorMap. All I need now is to add a legend or gradient item, which would represent the correspondence between the color and values I have on the faces.

Any Ideas how I can do it?

Thanks,

Igor.

[Luke Campagnola]

On Tue, Nov 25, 2014 at 3:17 PM, Igor Shovkun <igs...@gmail.com> wrote:

I have managed to plot and color it using pg.ColorMap. All I need now is to add a legend or gradient item, which would represent the correspondence between the color and values I have on the faces.

Any Ideas how I can do it?

It's not pretty, but there is a way to overlay 2D and 3D views:
https://groups.google.com/d/msg/pyqtgraph/PfJvmjIF3Dg/QVG9xUGk-zgJ

Does that work for you?

[Igor Shovkun]

I have already handled this problem myself. In case somebody wonders, I show the code:

import numpy as np

import pyqtgraph.opengl as gl

from pyqtgraph.Qt import QtCore, QtGui

import pyqtgraph as pg

pg.setConfigOption("useWeave", False) # this line eliminates some strange output lines

from pyqtgraph.pgcollections import OrderedDict

Gradients = OrderedDict([

('bw', {'ticks': [(0.0, (0, 0, 0, 255)), (1, (255, 255, 255, 255))], 'mode': 'rgb'}),

    ('hot', {'ticks': [(0.3333, (185, 0, 0, 255)), (0.6666, (255, 220, 0, 255)), (1, (255, 255, 255, 255)), (0, (0, 0, 0, 255))], 'mode': 'rgb'}),

    ('jet', {'ticks': [(1, (166, 0, 0, 255)), (0.32247191011235954, (0, 255, 255, 255)), (0.11348314606741573, (0, 68, 255, 255)), (0.6797752808988764, (255, 255, 0, 255)), (0.902247191011236, (255, 0, 0, 255)), (0.0, (0, 0, 166, 255)), (0.5022471910112359, (0, 255, 0, 255))], 'mode': 'rgb'}),

    ('summer', {'ticks': [(1, (255, 255, 0, 255)), (0.0, (0, 170, 127, 255))], 'mode': 'rgb'} ),

    ('space', {'ticks': [(0.562, (75, 215, 227, 255)), (0.087, (255, 170, 0, 254)), (0.332, (0, 255, 0, 255)), (0.77, (85, 0, 255, 255)), (0.0, (255, 0, 0, 255)), (1.0, (255, 0, 127, 255))], 'mode': 'rgb'}),

    ('winter', {'ticks': [(1, (0, 255, 127, 255)), (0.0, (0, 0, 255, 255))], 'mode': 'rgb'})

])

class Plotter:

def __init__(self):

pg.mkQApp()

# self.WindowNumber = 0

self.plot = None

self.plots = []

self.windows = []

def surfMesh(self, vertexes, faces, faceValues, cmap = 'space', drawEdges=False, smooth=False):

'''

plots data for the triangular mesh

Arguments:

float numpyarray vertexes: x-y coordinates of the vertices constituting the mesh

note: doesn't work with 3D. need some changes

int numpyarray faces: each element represents numbers of verticis constituting a face

float numpyarray data

str cmap:

'space' - default cmap developed by me. kinda cool :-)

'thermal'

'''

nbplt = len(self.plots)

# impose color scheme

if cmap in Gradients:

GradiendMode = Gradients[cmap]

else: raise TypeError('Color Map is not defined.')

# compute normalized value for color calculation

nbv = len(vertexes) # number of vertexes

xmin = faceValues.min()

xmax = faceValues.max()

if xmax != xmin:

x = (faceValues - xmin)/(xmax - xmin)

else: x = np.ones(len(faceValues)) * xmin

# Main window containing other widgets

win = QtGui.QWidget()

win.setWindowTitle('isqt plot')

self.windows.append(win)

# leyout windget controlling layour of other widgets

layout = QtGui.QGridLayout()

win.setLayout(layout)

# Widget - righthand side of the window allocated for the colorbar

cb = pg.GraphicsLayoutWidget()

# values axis

ax = pg.AxisItem('left')

if xmax != xmin:  # here we handle arrays with all values the same

# ax.setRange(np.round(xmin,2), np.round(xmax,2)) # to round numbers on ticks - doesn't work

ax.setRange(xmin, xmax)

else:

ax.setRange(xmin, xmax + 0.0001)

cb.addItem(ax)

# Gradient Editor widget

gw = pg.GradientEditorItem(orientation='right')

# load predefined color gradient

gw.restoreState(GradiendMode)

# left side of the window for 3D data representation

view = gl.GLViewWidget()

# it's basically 

view.setSizePolicy(cb.sizePolicy())

# add 3D widget to the left (first column)

layout.addWidget(view, 0, 0)

# add colorbar to the right (second column)

layout.addWidget(cb, 0, 1)

# Do not allow 2nd column (colorbar) to stretch

layout.setColumnStretch(1, 0)

# minimal size of the colorbar

layout.setColumnMinimumWidth(1, 120)

# Allow 1st column (3D widget) to stretch

layout.setColumnStretch(0, 1)

# horizontal size set to be large to prompt colormap to a minimum size

view.sizeHint = lambda: pg.QtCore.QSize(1700, 800)

cb.sizeHint = lambda: pg.QtCore.QSize(100, 800)

# this is to remove empty space between

layout.setHorizontalSpacing (0)

# set initial size of the window

win.resize(800,800)

win.show()

cb.addItem(gw)

cm = gw.colorMap()

# colors = cm.map(x)

colors = cm.mapToFloat(x)

# create 3D array to pass it to plotting function

vetrs3D = np.zeros([nbv,3])

if vertexes.shape == (nbv,3): vetrs3D = vertexes

else: 

print 'Converting 2D to 3D'

vetrs3D[:,:2] = vertexes

# compute distance to set initial camera pozition

dst = max(vertexes[:,0].max() - vertexes[:,0].min(), vertexes[:,1].max() - vertexes[:,1].min())

view.setCameraPosition(distance=dst)

plt = gl.GLMeshItem(vertexes=vetrs3D, faces=faces, faceColors=colors, drawEdges=drawEdges, smooth=smooth)

view.addItem(plt)

self.plots.append(plt)

And an example of using this stuff is:

verts = np.array([

    [0, 0, 0],

    [2, 0, 0],

    [1, 2, 0],

    [1, 1, 1],

])

faces = np.array([

    [0, 1, 2],

    [0, 1, 3],

    [0, 2, 3],

    [1, 2, 3]

])

values = np.array([0.0,2.0,3,20.0])

plt = Plotter()

plt.surfMesh(vertexes=verts, faces=faces, faceValues=values)

plt.show()

There might be some excessive variables in this script, that is due to the fact, this is only a part of the class I implemented for my plotting needs.

--
You received this message because you are subscribed to a topic in the Google Groups "pyqtgraph" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pyqtgraph/R-F_Co5XBJs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pyqtgraph+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pyqtgraph/CACZXET-sgjaHggE8On7a1baHR1sqtRS_xHv3FHfRUmKQr4gA0A%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

--

Best regards,

Igor.