skimage2.morphology.erosion

skimage2.morphology.erosion(image, footprint=None, *, out=None, mode='ignore', cval=0.0)

Return grayscale morphological erosion of an image.

Morphological erosion shrinks bright regions and enlarges dark regions. It assigns each pixel the minimum of the active neighborhood of that pixel. The values where footprint is 1 define this active neighborhood.

Parameters:

imagendarray

Input image.

footprintndarray or tuple, optional

The neighborhood expressed as a 2-D array of 1’s and 0’s. If None, use a cross-shaped footprint (so-called 1-connectivity). The footprint can also be provided as a sequence of smaller footprints as described in the notes below. See _Notes_ for more.

outndarray, optional

The array to store the result of the morphology. If None, a new array is allocated.

modestr, optional

The mode parameter determines how the array borders are handled. Valid modes are: ‘reflect’, ‘constant’, ‘nearest’, ‘mirror’, ‘wrap’, ‘max’, ‘min’, or ‘ignore’. If ‘max’ or ‘ignore’, pixels outside the image domain are assumed to be the maximum for the image’s dtype, which causes them to not influence the result. Default is ‘ignore’.

cvalscalar, optional

Value to fill past edges of input if mode is ‘constant’. Default is 0.0.

Returns:

outndarray, same shape and dtype as image

The result of the morphological erosion.

Notes

For uint8 (and uint16 up to a certain bit-depth) data, the lower algorithm complexity makes the skimage2.filters.rank.minimum() function more efficient for larger images and footprints.

The footprint can also be provided as a sequence of 2-tuples where the first element of each 2-tuple is a footprint ndarray and the second element is an integer describing the number of times it should be iterated. For example, footprint=[(np.ones((9, 1)), 1), (np.ones((1, 9)), 1)] would apply a 9x1 footprint followed by a 1x9 footprint resulting in a net effect that is the same as footprint=np.ones((9, 9)), but with lower computational cost. Most of the built-in footprints such as skimage2.morphology.disk() provide an option to automatically generate a footprint sequence of this type. Refer to the example Decompose flat footprints (structuring elements) for more insights.

If footprint contains even-sized dimensions, they are padded with zeros to an odd size at the front (at index 0) with pad_footprint().

Examples

>>> # Erosion shrinks bright regions
>>> import numpy as np
>>> from _skimage2.morphology import footprint_rectangle
>>> bright_square = np.array([[0, 0, 0, 0, 0],
...                           [0, 1, 1, 1, 0],
...                           [0, 1, 1, 1, 0],
...                           [0, 1, 1, 1, 0],
...                           [0, 0, 0, 0, 0]], dtype=np.uint8)
>>> erosion(bright_square, footprint_rectangle((3, 3)))
array([[0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0]], dtype=uint8)