The convex hull is defined by the smallest convex polygon that surrounds all foreground (white pixels) in the input image. The following code block demonstrates how to compute the convex hull for a binary image:
from skimage.morphology import convex_hull_imageim = rgb2gray(imread('../images/horse-dog.jpg'))threshold = 0.5im[im < threshold] = 0 # convert to binary imageim[im >= threshold] = 1chull = convex_hull_image(im)plot_images_horizontally(im, chull, 'convex hull', sz=(18,9))
The following screenshot shows the output:
The following code block plots the difference image of the original binary image and the computed ...
