I'm just testing very basic stuff about image manipulation with numpy ndarray data structure and matplotlib for show the image. I created a 2-dimensional array with np.zeros((n,m)) function where store my pixel, to keep things easy I'm just working in grayscale so I only need one value to represent my pixel. this is my code snippet:
import numpy as np
import matplotlib.pyplot as plt
matrix= np.zeros((4,4))
print(matrix)
plt.matshow(matrix, cmap=plt.cm.gray) #plot matrix
matrix+=128 #increase each pixel value by 128
print(matrix) #print the content of matrix
plt.matshow(matrix, cmap=plt.cm.gray) #plot matrix
What I expect from this code is that the first plot will be totally black because each "pixel" has value 0, and there's no problem here. Then I increase the pixel from 0 to 128 so the second plot should be totally grey, but is still black. Is like no operation is executed but matrix effectively has changed(checked from the print function).
So I tried to modify just some pixel "manually":
matrix[0,0]=255
matrix[0,3]=255
matrix[3,0]=255
matrix[3,3]=255
plt.matshow(matrix, cmap=plt.cm.gray) #plot matrix
and now 4 pixels are white. There are some aspects that I'm totally missing and I think that are related to numpy and some stuff about ndarray and how is managed.
There's anyone can explain me why this happens? thanks
