Moin moin, I'm using the find_boundaries function for separating labeled regions (separated by the watershed). Afterward I set the as True returned pixels to 0. This works quit well but since my objects (labeled regions) can be small (couple pixels) and seldom stick together I would like to limit the find_boundaries to connected regions. Is there a easy way to only find the boundaries not connected to 0. Or tell the watershed function to leave a space between separate objects? Example: import numpy as np from scipy import ndimage as nd from skimage.segmentation.boundaries import find_boundaries a = np.zeros((15,15)) a[4:9,4:9]=1 a[6:10,7:12]=2 a[1:3,1:3]=3 print "a" print a.astype(int) a[find_boundaries(a)]=0 print a.astype(int) Output: [[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 3 3 0 0 0 0 0 0 0 0 0 0 0 0] [0 3 3 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 1 1 1 1 1 0 0 0 0 0 0] [0 0 0 0 1 1 1 1 1 0 0 0 0 0 0] [0 0 0 0 1 1 1 2 2 2 2 2 0 0 0] [0 0 0 0 1 1 1 2 2 2 2 2 0 0 0] [0 0 0 0 1 1 1 2 2 2 2 2 0 0 0] [0 0 0 0 0 0 0 2 2 2 2 2 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]] After find_boundaries: [[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 3 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 1 1 1 1 0 0 0 0 0 0] [0 0 0 0 0 1 1 0 0 0 0 0 0 0 0] [0 0 0 0 0 1 1 0 2 2 2 2 0 0 0] [0 0 0 0 0 1 1 0 2 2 2 2 0 0 0] [0 0 0 0 0 0 0 0 2 2 2 2 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]] As you can see, the region 3 is almost gone. Thanks Siggi