This recipe demonstrates the use of the Python Imaging Library to apply a gaussian blur drop shadow to an image.
1 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 + abs(offset) + 2*border totalHeight = image.size + abs(offset) + 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) shadowTop = border + max(offset, 0) back.paste(shadow, [shadowLeft, shadowTop, shadowLeft + image.size, shadowTop + image.size] ) # 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) imageTop = border - min(offset, 0) back.paste(image, (imageLeft, imageTop)) return back if __name__ == "__main__": import sys image = Image.open(sys.argv) 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.