CODEWITHSUNDEEP
pythonimagenumpyimage-processingmatplotlib
Mourad Over Flow
Mourad Over Flow

Reputation: 191

Display pixels values as an image

So I have a 1280 × 960 .TIF format image. I have extracted all pixels which is 1.200.000 pixels.

After that, I have done some calculations using each pixel on the image, let's say :

result = each_pixel * some calculation

The result list has also 1.200.000 values.

Now what I am trying to do is to display two different images, the first with the original pixel values which means the original image, the second one with the result list, which means the result calculated for each pixel displayed in the exact same x and y coordinates of the original image.

How can I do that? I suppose that we can do it with something like the colormap but I am not sure.

Upvotes: 4

Views: 14167

Answers (2)

Tonechas
Tonechas

Reputation: 13723

You could reshape the one-dimensional array result in-place by changing its shape property:

import matplotlib.pyplot as plt
from skimage import io

image = io.imread('https://i.sstatic.net/5KUry.png')

each_pixel = image.ravel()
some_calculation = 64

result = each_pixel // some_calculation
result.shape = image.shape

fig, (ax0, ax1) = plt.subplots(1, 2)
ax0.imshow(image, cmap='gray')
ax0.axis('off')
ax0.set_title('image')
ax1.imshow(result, cmap='gray')
ax1.axis('off')
ax1.set_title('result')
plt.show()

lions

Upvotes: 3

mrk
mrk

Reputation: 10366

The matplotlib way of doing this would be:

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

plt.imsave('filename.png', np.array(result).reshape(1280,960), cmap=cm.gray)
plt.imshow(np.array(result).reshape(1280,960))

Note: You could also use glumpy

Upvotes: 5

Related Questions