[FIXED] Can a numpy image (W, H) with intensity be converted to rgb format (W, H, 3) and same display result as in matplotlib?

Issue

Can a numpy image (W, H) with intensity be converted to rgb format (W, H, 3) and same imshow result? Eg, a numpy image img1 = (128,128) with intensity (not greyscale) => convert it to img2 = (128,128,3). Running pyplot.imshow(img1) and pyplot.imshow(img2) can get the same display result.

Solution

matplotlib.imshow uses colormaps behind the scenes, and so can you.

1. Get the appropriate color map from matplotlib.colormaps (use "viridis" if you aren't sure)
2. Apply that to your input array

Here is the Colormap documentation in case you want to learn more.

Here is some sample code that shows how to accomplish it:

import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl

rng = np.random.default_rng(seed=42)

img1 = rng.random(size=(128, 128))
plt.imshow(img1)
plt.show()

img2 = mpl.colormaps['viridis'](img1)
plt.imshow(img2)
plt.show()

Answered By - BitsAreNumbersToo