Ufdjrw

With the imshow function from pyplot, I can plot an (M,N) size image, where each value in the (M,N) sized array is associated to a color from a colormap

import numpy as np
import matplotlib.pyplot as plt

A = np.random.rand(5,5)
plt.imshow(A, cmap='jet')
plt.colorbar()
plt.show()

I have an (2,M,N) sized array (let's call it B) that I would like to show. B[0] gives the hue and B[1] is the alpha. Here, B[0] would be the same as A in the first example.

I would like to write something like plt.imshow(B[0], cmap='jet', alpha=B[1]). But the value of the alpha has to be the same for the entire image, since the alpha option has to be a scalar or None

With imshow, I can also specify an (M,N,3) or an (M,N,4) array, where the value for R,G,B and alpha are given. However, I don't know how to get the RGB values from my initial (M,N) sized array A.

I tried setting the color with the HSV scheme from my values, then converting to RGB and plotting with those colors

from matplotlib import colors

B = np.random.rand(2,5,5)

hue = B[0]
sat = B[1]
val = np.ones(B[0].shape)

rgb = colors.hsv_to_rgb(np.array([hue, sat, val]).T)

plt.imshow(rgb)
plt.show()

But then, I don't know how to make a color bar from the actual values in B.

I'm going for something like this:

where the vertical part of the colorbar shows the value in B[0] and the horizontal part of the colorbar shows the value in B[1].

The first part is relatively straight forward:

import numpy as np
import matplotlib.pyplot as plt

A = np.random.rand(5,5)
cmap = plt.get_cmap('jet')

rgba_img = cmap(A)
rgba_img[:, :, 3] = np.random.rand(5, 5)

plt.imshow(rgba_img)
plt.colorbar()
plt.show()

Making a colorbar is more difficult. You will basically need to replace the pcolor that is used to make the colorbar with another one that looks as you want it to.