Better DPC Visualization

[CSSFrancis]

Is your feature request related to a problem? Please describe.
@magnunor do you have any suggestions for how to better visualize DPC measurements in pyxem? It would be nice to have something like:

Where we could overlay some displacement on the navigator. @ericpre Any thoughts as well?

Describe the solution you'd like
It looks like most of the color wheels are some version of https://matplotlib.org/stable/api/collections_api.html#matplotlib.collections.QuadMesh which means this should work out quite nicely :).

[CSSFrancis]

For example:

import hyperspy.api as hs
import numpy as np


n = 360  #the number of secants for the mesh
t = np.linspace(0,2*np.pi,n)   #theta values
r = np.linspace(.6,1,2)        #radius values change 0.6 to 0 for full circle

y= np.sin(t[:,np.newaxis])*r[np.newaxis,:]/10
x= np.cos(t[:,np.newaxis])*r[np.newaxis,:]/10
coords = np.stack([x, y], axis=-1)

quad= hs.plot.markers.Markers(collection=matplotlib.collections.QuadMesh, coordinates=coords,
                              array=t[1:], cmap="hsv",
                              transform="axes",offset_transform="axes", 
                              offsets=[[.8,.8]], alpha=1)
                              
                              
s = hs.signals.Signal2D(np.ones((100,100,10,10)))

im = np.random.randint(1,360,(100,100))
nav = hs.signals.Signal2D(im)

s.plot(navigator=nav, navigator_kwds=dict(cmap="hsv", colorbar=False))

s.add_marker(quad, plot_on_signal=False)

[CSSFrancis]

Just playing around with other things...

import hyperspy.api as hs
import numpy as np


t = np.linspace(0,2*np.pi,n)   #theta values
r = np.linspace(.6,1,2)        #radius values change 0.6 to 0 for full circle

y= np.linspace(0,1, 11)
x= np.linspace(0,1, 11)

xx,yy = np.meshgrid(x,y)

coords = np.stack([xx, yy], axis=-1)
t = np.zeros((10,10))
t[t==0]=np.nan
rr,cc =disk((5,5), radius=5)
t[rr, cc]=np.random.rand(len(rr))

quad= hs.plot.markers.Markers(collection=matplotlib.collections.QuadMesh, coordinates=coords,
                              array=t, cmap="hsv",
                          transform="axes",offset_transform="display", 
                              offsets=[[0,0]], alpha=0.8)
                              
s = hs.signals.Signal2D(np.random.rand(100,100,10,10))

im = np.random.randint(1,360,(100,100))
nav = hs.signals.Signal2D(im)

s.plot( navigator_kwds=dict(cmap="hsv", colorbar=False))

s.add_marker(quad, plot_on_signal=True)

[magnunor]

Agreed! This is really nice. Similar functionality exist with in the
DPCSignal.get_color_image_with_indicator` method: https://github.com/pyxem/pyxem/blob/main/pyxem/signals/differential_phase_contrast.py#L568

However, it is a a bit sub-optimal, as it returns a matplotlib figure.

The overlay can be added to DPCSignal.get_color_signal: https://github.com/pyxem/pyxem/blob/main/pyxem/signals/differential_phase_contrast.py#L495

---

The color wheel itself is made here: https://github.com/pyxem/pyxem/blob/main/pyxem/utils/pixelated_stem_tools.py#L214