3. Adaptive RGB decorator

There are many filters that are designed to work with gray-scale images but not with color images. To simplify the process of creating functions that can adapt to RGB images, scikit-image provides the adapt_rgb decorator.

We will demonsrate the use with the famous Sobel filter.

To actually use the adapt_rgb decorator, you have to decide how you want to adapt the RGB image for use with the gray-scale filter. There are two pre-defined handlers:

Pass each of the RGB channels to the filter one-by-one, and stitch the results back into an RGB image.

Convert the RGB image to HSV and pass the value channel to the filter. The filtered result is inserted back into the HSV image and converted back to RGB.

3.1. Wrap filter functions with decorators

from skimage.color.adapt_rgb import adapt_rgb, each_channel, hsv_value
from skimage import filters

def sobel_each(image):
    return filters.sobel(image)

def sobel_hsv(image):
    return filters.sobel(image)

3.2. We can use these functions as we would normally use them, but now they work with both gray-scale and color images.

from skimage import data
from skimage.exposure import rescale_intensity
import matplotlib.pyplot as plt
image = data.astronaut()
(512, 512, 3)

3.3. Normal Sobel filter does not work on this image

except ValueError as e:

3.4. When used with the decorator, it works


3.5. We can take the complement and rescale the intensity

fig, (ax_each, ax_hsv) = plt.subplots(ncols=2, figsize=(14, 7))

# We use 1 - sobel_each(image) but this won't work if image is not normalized
ax_each.imshow(rescale_intensity(1 - sobel_each(image)))
ax_each.set_xticks([]), ax_each.set_yticks([])
ax_each.set_title("Sobel filter computed\n on individual RGB channels")

# We use 1 - sobel_hsv(image) but this won't work if image is not normalized
ax_hsv.imshow(rescale_intensity(1 - sobel_hsv(image)))
ax_hsv.set_xticks([]), ax_hsv.set_yticks([])
ax_hsv.set_title("Sobel filter computed\n on value converted image (HSV)")
Text(0.5, 1.0, 'Sobel filter computed\n on value converted image (HSV)')