How to convert a grayscale image into a list of pixel values?

Question

I am trying to create a python program which takes a grayscale, 24*24 pixel image file (I haven't decided on the type, so suggestions are welcome) and converts it to a list of pixel values from 0 (white) to 255 (black).

I plan on using this array for creating a MNIST -like bytefile of the picture, that can be recognized by Tensor-Flow handwriting recognition algorithms.

I have found the Pillow library to be the most useful in this task, by iterating over each pixel and appending its value to an array

from PIL import Image

img = Image.open('eggs.png').convert('1')
rawData = img.load()
data = []
for y in range(24):
    for x in range(24):
        data.append(rawData[x,y])

Yet this solution has two problems:

1. The pixel values are not stored as integers, but pixel objects which cannot be further mathematically manipulated and are therefore useless.
2. Even the Pillow docs state that:

   Accessing individual pixels is fairly slow. If you are looping over all of the pixels in an image, there is likely a faster way using other parts of the Pillow API.

Answer 1

You can convert the image data into a Python list (or list-of-lists) like this:

from PIL import Image

img = Image.open('eggs.png').convert('L')  # convert image to 8-bit grayscale
WIDTH, HEIGHT = img.size

data = list(img.getdata()) # convert image data to a list of integers
# convert that to 2D list (list of lists of integers)
data = [data[offset:offset+WIDTH] for offset in range(0, WIDTH*HEIGHT, WIDTH)]

# At this point the image's pixels are all in memory and can be accessed
# individually using data[row][col].

# For example:
for row in data:
    print(' '.join('{:3}'.format(value) for value in row))

# Here's another more compact representation.
chars = '@%#*+=-:. '  # Change as desired.
scale = (len(chars)-1)/255.
print()
for row in data:
    print(' '.join(chars[int(value*scale)] for value in row))

Here's an enlarged version of a small 24x24 RGB eggs.png image I used for testing:

Here's the output from the first example of access:

And here the output from the second example:

@ @ % * @ @ @ @ % - . * @ @ @ @ @ @ @ @ @ @ @ @
@ @ .   . + @ # .     = @ @ @ @ @ @ @ @ @ @ @ @
@ *             . .   * @ @ @ @ @ @ @ @ @ @ @ @
@ #     . .   . .     + % % @ @ @ @ # = @ @ @ @
@ %       . : - - - :       % @ % :     # @ @ @
@ #     . = = - - - = - . . = =         % @ @ @
@ =     - = : - - : - = . .     . : .   % @ @ @
%     . = - - - - : - = .   . - = = =   - @ @ @
=   .   - = - : : = + - : . - = - : - =   : * %
-   .   . - = + = - .   . - = : - - - = .     -
=   . : : . - - .       : = - - - - - = .   . %
%   : : .     . : - - . : = - - - : = :     # @
@ # :   .   . = = - - = . = + - - = - .   . @ @
@ @ #     . - = : - : = - . - = = : . .     # @
@ @ %     : = - - - : = -     : -   . . .   - @
@ @ *     : = : - - - = .   . - .   .     . + @
@ #       . = - : - = :     : :   .   - % @ @ @
*     . . . : = = - : . .   - .     - @ @ @ @ @
*   . .       . : .   . .   - = . = @ @ @ @ @ @
@ :     - -       . . . .     # @ @ @ @ @ @ @ @
@ @ = # @ @ *     . .     . - @ @ @ @ @ @ @ @ @
@ @ @ @ @ @ @ .   .   . # @ @ @ @ @ @ @ @ @ @ @
@ @ @ @ @ @ @ -     . % @ @ @ @ @ @ @ @ @ @ @ @
@ @ @ @ @ @ @ # . : % @ @ @ @ @ @ @ @ @ @ @ @ @

Access to the pixel data should now be faster than using the object img.load() returns (and the values will be integers in the range of 0..255).

Answer 2

You can access the greyscale value of each individual pixel by accessing the r, g, or b value, which will all be the same for a greyscale image.

Ie

img = Image.open('eggs.png').convert('1')
rawData = img.load()
data = []
for y in range(24):
    for x in range(24):
        data.append(rawData[x,y][0])

This doesn't solve the problem of access speed.

I'm more familiar with scikit-image than Pillow. It seems to me that if all you are after is listing the greyscale values, you could use scikit-image, which stores images as numpy arrays, and use img_as_ubyte to represent the image as a uint array, containing values between 0 and 255.

Images are NumPy Arrays provides a good starting point to see what the code looks like.