I have to say I found the ImageView example a bit too complex to be a
useful demo of the imageView() for me when I was learning this. Steep
learning curve.
Basically what everyone else has said so far- the key is the input image
array.
The 2d arrays should ideally be stacked into a 3+ dimensional array, which
you can then slice to extract whichever single or multiple channels you
want.
I used a pg.ImageView instance, assigning the volume with
.setImage(my_img_nparray)
This shows a horizontal slider which you can use to navigate through your z
axis slices. This (I think) shows whenever you load a 3d volume. Set the
pixel data type (monochrome/rgba) with the levelMode param in .setImage()
or when creating the imageView() object.
There is a monochrome example of the slider in the ‘Data Slicing’ example.
I have rewritten it as RGB[A] and have cut down the raw example as it was a
little more complex than needed.
I imagine that you'll need to merge the different channels as necessary
into a single numpy array - I have tried to suggest one way* of doing it
with numpy in the code. I am always surprised how fast this is with numpy.
(*Not necessarily the best way, or even a good way- I am not a professional
developer)
Also I could never remember which way round the x, y, and z planes went in
the numpy array so I always just went by trial and error. Please don't go
by the x, y and z in the example- that is from the original Data Slicer
code and is not mine.
Hope this helps.
"""
Data Slicer demo cut down and rewritten to be RGB[A] rather than monochrome.
"""
import numpy as np
import pyqtgraph as pg
from pyqtgraph.Qt import QtWidgets
app = pg.mkQApp("RGB[A] Data Slicing Example")
## Create window with two ImageView widgets
win = QtWidgets.QMainWindow()
win.resize(800,800)
imv = pg.ImageView()
win.setCentralWidget(imv)
win.show()
x1 = np.linspace(-30, 10, 128)[:, np.newaxis, np.newaxis]
x2 = np.linspace(-20, 20, 128)[:, np.newaxis, np.newaxis]
y = np.linspace(-30, 10, 128)[np.newaxis, :, np.newaxis]
z = np.linspace(-20, 20, 128)[np.newaxis, np.newaxis, :]
d1 = np.sqrt(x1**2 + y**2 + z**2)
d2 = 2*np.sqrt(x1[::-1]**2 + y**2 + z**2)
d3 = 4*np.sqrt(x2**2 + y[:,::-1]**2 + z**2)
# Create R, G and B channels
data = np.empty([128, 128, 128, 3]) # alternatively you can stack the
arrays with numpy to make the 4th axis
data[:, :, :, 0] = (np.sin(d1) / d1**2)
data[:, :, :, 1] = (np.sin(d2) / d2**2)
data[:, :, :, 2] = (np.sin(d3) / d3**2)
## Display the data
imv.setImage(data, levelMode='rgba')
imv.setLevels(-0.003, 0.003)
# Hide the stuff that gets in the way
imv.ui.histogram.hide()
imv.ui.roiBtn.hide()
imv.ui.menuBtn.hide()
if __name__ == '__main__':
pg.exec()
On Thu, 20 Jun 2024 at 10:25 PM, Martin Chase <[email protected]>
wrote:
> Hey Jason,
>
> Have a look at `python -m pyqtgraph.examples.ImageView`.
> I know 3-channel is really easy: each of red, green and blue would be
> assigned to a different channel. The expected format for that would be [z
> position][col][row][channel] (row and col can be switched via
> `pg.setConfigOptions(imageAxisOrder="row-major", and in fact performs
> better that way)`. You could do 4 channels by assigning them to CMYK (and
> then transforming it to RGB; PIL will do this, or the math is easy enough
> to do by hand/numpy)? I've never tried this, so I don't know how well it
> would work in practice, but it should be easy enough.
>
> --
> You received this message because you are subscribed to the Google Groups
> "pyqtgraph" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pyqtgraph/CAD_p8v2gwr81GY9FPA%3DoBXcmgsyp8g0oyUbK%2BbXe7ZiCvu8PPA%40mail.gmail.com
> <https://groups.google.com/d/msgid/pyqtgraph/CAD_p8v2gwr81GY9FPA%3DoBXcmgsyp8g0oyUbK%2BbXe7ZiCvu8PPA%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
>
--
You received this message because you are subscribed to the Google Groups
"pyqtgraph" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/pyqtgraph/CALWkJ8TvM-HDUoeUVg1h9tSrcUTewpvy1rsX6vU%2BQPKWr4dOdA%40mail.gmail.com.
