skimage2.morphology.convex_hull_object

skimage2.morphology.convex_hull_object(image, *, connectivity=2)

Compute the convex hull image of individual objects in a binary image.

The convex hull is the set of pixels included in the smallest convex polygon that surround all white pixels in the input image.

Parameters:

imagendarray of shape (M, N)

Binary input image.

connectivity{1, 2}, int, optional

Determines the neighbors of each pixel. Adjacent elements within a squared distance of connectivity from pixel center are considered neighbors.:

1-connectivity      2-connectivity
      [ ]           [ ]  [ ]  [ ]
       |               \  |  /
 [ ]--[x]--[ ]      [ ]--[x]--[ ]
       |               /  |  \
      [ ]           [ ]  [ ]  [ ]

Returns:

hullndarray of bool

Binary image with pixels inside convex hull set to True.

Notes

This function uses skimage.morphology.label to define unique objects, finds the convex hull of each using convex_hull_image, and combines these regions with logical OR. Be aware the convex hulls of unconnected objects may overlap in the result. If this is suspected, consider using convex_hull_image separately on each object or adjust connectivity.