How do you composite an image onto another image with PIL in Python?

Question

I need to take an image and place it onto a new, generated white background in order for it to be converted into a downloadable desktop wallpaper. So the process would go:

Generate new, all white image with 1440x900 dimensions
Place existing image on top, centered
Save as single image

In PIL, I see the ImageDraw object, but nothing indicates it can draw existing image data onto another image. Suggestions or links anyone can recommend?

FYI, [difference between Image.blend, paste, composite and alpha_composite in Pillow](https://stackoverflow.com/q/72362114/5101148) — Simba, May 27 '22 at 17:09

unutbu · Accepted Answer · 2018-03-09T10:14:27.870

189

This can be accomplished with an Image instance's paste method:

from PIL import Image
img = Image.open('/path/to/file', 'r')
img_w, img_h = img.size
background = Image.new('RGBA', (1440, 900), (255, 255, 255, 255))
bg_w, bg_h = background.size
offset = ((bg_w - img_w) // 2, (bg_h - img_h) // 2)
background.paste(img, offset)
background.save('out.png')

This and many other PIL tricks can be picked up at Nadia Alramli's PIL Tutorial

edited Mar 09 '18 at 10:14

answered Apr 01 '10 at 21:42

unutbu

842,883
184
1,785
1,677

1

You may need to import depending on your module/system/version: from PIL import Image – Nuno Aniceto Apr 24 '15 at 10:57
3

Thanks @NunoAniceto, I've changed it to `from PIL import Image` to make the code more [compatible with Pillow](http://pillow.readthedocs.org/en/latest/porting-pil-to-pillow.html). – unutbu Apr 24 '15 at 11:46
If you're using Python 3.x, check https://stackoverflow.com/a/17530159/7326714 to fix the 'offeset' tuple integer error. – LucSpan Mar 09 '18 at 10:04

Martin Thoma · Answer 2 · 2016-11-23T10:50:57.960

Based on unutbus answer:

#!/usr/bin/env python

from PIL import Image
import math


def resize_canvas(old_image_path="314.jpg", new_image_path="save.jpg",
                  canvas_width=500, canvas_height=500):
    """
    Place one image on another image.

    Resize the canvas of old_image_path and store the new image in
    new_image_path. Center the image on the new canvas.
    """
    im = Image.open(old_image_path)
    old_width, old_height = im.size

    # Center the image
    x1 = int(math.floor((canvas_width - old_width) / 2))
    y1 = int(math.floor((canvas_height - old_height) / 2))

    mode = im.mode
    if len(mode) == 1:  # L, 1
        new_background = (255)
    if len(mode) == 3:  # RGB
        new_background = (255, 255, 255)
    if len(mode) == 4:  # RGBA, CMYK
        new_background = (255, 255, 255, 255)

    newImage = Image.new(mode, (canvas_width, canvas_height), new_background)
    newImage.paste(im, (x1, y1, x1 + old_width, y1 + old_height))
    newImage.save(new_image_path)

resize_canvas()

Remember to use Pillow (Documentation, GitHub, PyPI) instead of python-imaging as Pillow works with Python 2.X and Python 3.X.

score 5 · Answer 3 · answered Mar 05 '17 at 08:46

This is to do something similar

Where I started was by generating that 'white background' in photoshop and exporting it as a PNG file. Thats where I got im1 (Image 1). Then used the paste function cause it's way easier.

from PIL import Image

im1 = Image.open('image/path/1.png')
im2 = Image.open('image/path/2.png')
area = (40, 1345, 551, 1625)  
im1.paste(im2, area)
                   l>(511+40) l>(280+1345)
         |    l> From 0 (move, 1345px down) 
          -> From 0 (top left, move 40 pixels right)

Okay so where did these #'s come from? (40, 1345, 551, 1625) im2.size (511, 280) Because I added 40 right and 1345 down (40, 1345, 511, 280) I must add them to the original image size which = (40, 1345, 551, 1625)

im1.show()

to show your new image

score 0 · Answer 4 · edited Apr 17 '21 at 06:11

0

Maybe too late, but for such image operations, we do use ImageSpecField in model with original image.

edited Apr 17 '21 at 06:11

Jenia

374
1
4
15

answered Oct 30 '13 at 09:34

profuel

128
5

score 0 · Answer 5 · answered Apr 01 '10 at 21:41

0

Image.blend()? [link]

Or, better yet, Image.paste(), same link.

answered Apr 01 '10 at 21:41

Felix

88,392
43
149
167

"Creates a new image by interpolating between the given images, using a constant alpha. Both images must have the same size and mode." From the documentation, it appears they can't be different sizes. – Sebastian Apr 01 '10 at 21:44
I noted `Image.paste()`, too, which was ultimately the solution. – Felix Apr 02 '10 at 10:50
2

The url expired – Phani Rithvij Nov 24 '19 at 12:06

score 0 · Answer 6 · answered Mar 06 '23 at 06:06

Try this code below

from PIL import Image

# Load the existing image
existing_image = Image.open("existing_image.png")

# Create a new white image with 1440x900 dimensions
new_image = Image.new("RGB", (1440, 900), (255, 255, 255))

# Calculate the center position for the existing image
x = (new_image.width - existing_image.width) // 2
y = (new_image.height - existing_image.height) // 2

# Paste the existing image onto the new white image at the center position
new_image.paste(existing_image, (x, y))

# Save the new image as a PNG file
new_image.save("new_wallpaper.png")

How do you composite an image onto another image with PIL in Python?

6 Answers6

Linked