Work with PNG binary images in PIL / pillow

Question

When converting PNG binary files from a PIL image object to a numpy array, the values ​​are the same regardless of whether the original image is inverted or not.

For example, both of these images generate identical numpy arrays.

import numpy as np
from PIL import Image
t = Image.open('t.png')
t_inverted = Image.open('t_inverted.png')
np.asarray(t)
np.asarray(t_inverted)

output either np.asarray(t)or np.asarray(t_inverted):

array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
       [1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
       [1, 0, 1, 1, 0, 0, 1, 1, 0, 1],
       [1, 1, 1, 1, 0, 0, 1, 1, 1, 1],
       [1, 1, 1, 1, 0, 0, 1, 1, 1, 1],
       [1, 1, 1, 1, 0, 0, 1, 1, 1, 1],
       [1, 1, 1, 1, 0, 0, 1, 1, 1, 1],
       [1, 1, 1, 1, 0, 0, 1, 1, 1, 1],
       [1, 1, 1, 0, 0, 0, 0, 1, 1, 1],
       [1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=uint8)

I expected 0s and 1s to be inverted. Why are they the same?

Answer 1

Both of these PNG files are indexed. They contain the same data array, but only the values ​​0 and 1 that you see, but these values ​​are not intended for pixel colors. They must be indices in the palette. In the first file, the palette

 Index     RGB Value
   0    [  0,   0,   0]
   1    [255, 255, 255]

and in the second file, the palette

 Index     RGB Value
   0    [255, 255, 255]
   1    [  0,   0,   0]

, Image numpy, .

, convert() Image RGB:

t = Image.open('t.png')
t_rgb = t.convert(mode='RGB')
arr = np.array(t_rgb)