Expand segmentation labels without overlapΒΆ

Given several connected components represented by a label image, these connected components can be expanded into background regions using skimage.segmentation.expand_labels(). In contrast to skimage.morphology.dilation() this method will not let connected components expand into neighboring connected components with lower label number.

Sobel+Watershed, Expanded labels
import numpy as np
import matplotlib.pyplot as plt

from skimage.filters import sobel
from skimage.measure import label
from skimage.segmentation import watershed, expand_labels
from skimage.color import label2rgb
from skimage import data

coins = data.coins()

# Make segmentation using edge-detection and watershed.
edges = sobel(coins)

# Identify some background and foreground pixels from the intensity values.
# These pixels are used as seeds for watershed.
markers = np.zeros_like(coins)
foreground, background = 1, 2
markers[coins < 30.0] = background
markers[coins > 150.0] = foreground

ws = watershed(edges, markers)
seg1 = label(ws == foreground)

expanded = expand_labels(seg1, distance=10)

# Show the segmentations.
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(9, 5),
                         sharex=True, sharey=True)

color1 = label2rgb(seg1, image=coins, bg_label=0)

color2 = label2rgb(expanded, image=coins, bg_label=0)
axes[1].set_title('Expanded labels')

for a in axes:

Total running time of the script: ( 0 minutes 0.807 seconds)

Gallery generated by Sphinx-Gallery