This recipe demonstrates the use of the Python Imaging Library to apply a gaussian blur drop shadow to an image.
Download
Copy to clipboard1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 | """
Drop shadows with PIL.
Author: Kevin Schluff
License: Python license
"""
from PIL import Image, ImageFilter
def dropShadow( image, offset=(5,5), background=0xffffff, shadow=0x444444,
border=8, iterations=3):
"""
Add a gaussian blur drop shadow to an image.
image - The image to overlay on top of the shadow.
offset - Offset of the shadow from the image as an (x,y) tuple. Can be
positive or negative.
background - Background colour behind the image.
shadow - Shadow colour (darkness).
border - Width of the border around the image. This must be wide
enough to account for the blurring of the shadow.
iterations - Number of times to apply the filter. More iterations
produce a more blurred shadow, but increase processing time.
"""
# Create the backdrop image -- a box in the background colour with a
# shadow on it.
totalWidth = image.size[0] + abs(offset[0]) + 2*border
totalHeight = image.size[1] + abs(offset[1]) + 2*border
back = Image.new(image.mode, (totalWidth, totalHeight), background)
# Place the shadow, taking into account the offset from the image
shadowLeft = border + max(offset[0], 0)
shadowTop = border + max(offset[1], 0)
back.paste(shadow, [shadowLeft, shadowTop, shadowLeft + image.size[0],
shadowTop + image.size[1]] )
# Apply the filter to blur the edges of the shadow. Since a small kernel
# is used, the filter must be applied repeatedly to get a decent blur.
n = 0
while n < iterations:
back = back.filter(ImageFilter.BLUR)
n += 1
# Paste the input image onto the shadow backdrop
imageLeft = border - min(offset[0], 0)
imageTop = border - min(offset[1], 0)
back.paste(image, (imageLeft, imageTop))
return back
if __name__ == "__main__":
import sys
image = Image.open(sys.argv[1])
image.thumbnail( (200,200), Image.ANTIALIAS)
dropShadow(image).show()
dropShadow(image, background=0xeeeeee, shadow=0x444444, offset=(0,5)).show()
|
Adding a gaussian blur drop shadow is an easy way to give a flat image a natural looking 3D effect.
The dropShadow() function given here uses PIL's built in BLUR filter to produce a blurred shadow. Since this filter uses a small kernel (5x5), it must be applied a number of times to give a smooth blur at the edges of the shadow. As a result, processing of large images or multiple small images can be fairly slow.
One possible speedup is to cache the backdrop (shadow+background) images. An application, such as a web thumbnail gallery, that has many images of the same size could reuse these backdrops without re-applying the filter.
Other possible enhancments include support for non-rectangular images, like text, or the use of an alpha channel rather than a solid backgound for a semi-transparent shadow.
Image as a background, not a colour. How would I do to make the background an image rather than just a background colour?
Thanks for the code, helped alot. I was running into problems with putting a drop shadow on rectangles with rounded edges. To get it to work I used a mask when pasting the image. Code looks like this: # Paste the input image onto the shadow backdrop
imageLeft = border - min(offset[0], 0) imageTop = border - min(offset[1], 0) back.paste(image, (imageLeft, imageTop), image)#my change here return back
back.paste(image, (imageLeft, imageTop), image)#sorry that was ugly this is the line that I changed
Hi, Could anybody please me some suggestions on adding shadows to the object/objects inside an image?. How to modify the above program for non rectangular boundaries?
Thanks, Ameet.