19. ImageFilter Kernel
19.1. Kernel
Use the
Image.filter(ImageFilter.Kernel(size, kernel, scale=None, offset=0) method to apply a convolution to an image.- ImageFilter.Kernel(size, kernel, scale=None, offset=0)
- Parameters:size - Kernel size, given as (width, height). This must be (3, 3) or (5, 5).kernel - A sequence containing kernel weights (integer or floating point).e.g. (-1, -1, -1, -1, 9, -1, -1, -1, -1)scale - Scale factor. If given, the result for each pixel is divided by this value. The default is the sum of the kernel weights.offset - Offset. If given, this value is added to the result, after it has been divided by the scale factor.
Returns type: An image.
from PIL import Image, ImageFilter
with Image.open("test_images/river_valley.jpg") as im:
new_im = im.filter(ImageFilter.Kernel((3, 3), (-1, -1, -1, -1, 9, -1, -1, -1, -1), 1, 0))
new_im.save("filters/Kernel.png")
new_im = im.filter(ImageFilter.Kernel((3, 3), (0, -1, 0, -1, 5, -1, 0, -1, 0), 1, 0))
new_im.save("filters/Kernel2.png")
