CODEWITHSUNDEEP
pythonperformancenumpy
Deepesh
Deepesh

Reputation: 99

Apply color map to grayscale image to generate RGB images efficiently

Is there an efficient way of applying color map dictionary to grayscale image to convert to RGB image using numpy functions?

For eg. I have a a grayscale image as numpy array . 

grayscale_image = array([[0., 0., 3.],
       [0., 2., 0.]])

and a color map like 

color_map = {3: (1,2,3), 2: (4,5,6)}

How can I generate RGB image like 

rgb_image = 
array([[[0., 0., 0.],
        [0., 0., 0.],
        [1., 2., 3.]],
       [[0., 0., 0.],
        [4., 5., 6.],
        [0., 0., 0.]]])

Upvotes: 2

Views: 1317

Answers (1)

Mark
Mark

Reputation: 92461

You can take advantage of numpy's very convenient indexing if you make your color map an array instead of a dictionary. If you have 256 shades of gray, you will have a color map of shape [256, 3]. Then you can directly index:

import numpy as np

gray = np.array([
    [0, 0, 3],
    [0, 2, 0]
])

color_map = np.array([
    [0,0,0],
    [0,0,0],
    [4,5,6],
    [1,2,3], 
    # ... remaining color map values
])

rgb = color_map[gray]

Result:

array([[[0, 0, 0],
        [0, 0, 0],
        [1, 2, 3]],

       [[0, 0, 0],
        [4, 5, 6],
        [0, 0, 0]]])

Upvotes: 3

Related Questions