Question

I’m using PIL for a uni project and we have one task where we have to darken or brighten an image without using any of PIL’s functions to do so. The function takes the original filename, the action (‘lighten’ or ‘darken’) and the extent (in percent – an int between 0 and 100). Here’s what I’ve come up with so far:

from PIL import Image

def change_brightness(filename, action, extent):
    """
    This function either increases or decreases the brightness of an image
    by altering each pixel in each band

    """
    #load the original image into a list
    original_image = Image.open(filename, 'r')
    pixels = original_image.getdata()

    #initialise the new image
    new_image = Image.new('RGB', original_image.size)
    new_image_list = []

    brightness_multiplier = 1.0

    if action == 'lighten':
        brightness_multiplier += (extent/100)
    else:
        brightness_multiplier -= (extent/100)

    #for each pixel, append the brightened or darkened version to the new image list
    for pixel in pixels:
        new_pixel = (int(pixel[0] * brightness_multiplier),
                     int(pixel[1] * brightness_multiplier),
                     int(pixel[2] * brightness_multiplier))

        #check the new pixel values are within rgb range
        for pixel in new_pixel:
            if pixel > 255:
                pixel = 255
            elif pixel < 0:
                pixel = 0

        new_image_list.append(new_pixel)

    #save the new image
    new_image.putdata(new_image_list)
    new_image.save('colour_brightness.jpg')

When I run this, the new image is not modified from the original (save from some new jpg artefacts). I tried brightness_multiplier with an explicit value (1.1 for lighten, and 0.9 for darken) and it worked, so I have no idea why it isn’t working when I have it taking the value from the extent argument.

If anyone could shed some light it would be greatly appreciated!

Answer 1

It’s an issue of integer division in the (extent/100) expression. To rectify this, you can:

use a floating point literal

extent/100.0

Handy if a term is a literal.

convert either the numerator or the denominator into float

float(extent)/a_hundred

If no term is a literal.

make sure that / is floating-point division

from __future__ import division

Insert that at the beginning of your source file, like all __future__ statements.

invoke python with -Qnew

python -Qnew

If using python2, otherwise

use python3 🙂

In all cases, // remains integer division.