Get indices for all elements adjacent to zeros in a numpy 2d array

4
votes

I have a set of 2d arrays that hold latitudes and longitudes for a map area, with a binary mask that denotes land (0) or ocean (1). What I am interested in doing is extracting the indices for all coastal ocean elements (mask elements 1 that are adjacent to a 0, including diagonally) so that I can use those indices to extract the lats+lons of all coastal elements from the other arrays.

Given an array:

a = np.array([[1, 1, 1, 1, 1],
[1, 1, 0, 1, 1],
[1, 0, 0, 0, 1],
[1, 1, 0, 1, 1],
[1, 1, 1, 1, 1]])

I'm looking for a way to return:

(array([0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4]), 
array([1, 2, 3, 0, 1, 3, 4, 0, 4, 0, 1, 3, 4, 1, 2, 3]))

where each array has the indices for each axis.

This output format is similar to np.where(), I guess what I'm trying to do is np.where(a == adjacent to 0).

1
pythonarraysnumpy

1 Answers

3
votes

Let's try convolve2d:

from scipy.signal import convolve2d

kernel = np.full((3,3), 1)

# remove center of kernel -- not count 1 at the center of the square
# we may not need to remove center
# in which case change the mask for counts
kernel[1,1]=0

# counts 1 among the neighborhoods
counts = convolve2d(a, kernel, mode='same', 
                    boundary='fill', fillvalue=1)

# counts==8 meaning surrounding 8 neighborhoods are all 1
# change to 9 if we leave kernel[1,1] == 1
# and we want ocean, i.e. a==1
np.where((counts != 8) & (a==1))

Output:

(array([0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4]),
 array([1, 2, 3, 0, 1, 3, 4, 0, 4, 0, 1, 3, 4, 1, 2, 3]))