masking a python 2d array to change values in a circle under a certain threshold

user_17320191 0
Viewed 195

As the title says, I'm trying to change the values of a circle in a 2D array if they fulfill a certain condition. So far I've got this from a different question on here:

import numpy as np

x = np.arange(0, 20)
y = np.arange(0, 20)
arr = np.zeros((y.size, x.size))

cx = 10.
cy = 10.
r = 4.

mask = (x[np.newaxis, :] - cx) ** 2 + (y[:, np.newaxis] - cy) ** 2 < r ** 2
arr[mask] = 1

cx and cy are the coordinates of the circle's center, r is its radius. mask contains the circle, and all of the elements in it are set to 1, which works fine. What I want to do is only change their value only if they are less than 5 or whatever some similar condition. Obviously this array only has 0s, but my real data would have other values set beforehand).

mask boolean numpy arrays python
1 Answers

Masks are booleans, and you can combine them with other masks using bitwise operations or logical operations:

circle_mask = (x[np.newaxis, :] - cx) ** 2 + (y[:, np.newaxis] - cy) ** 2 < r ** 2
less_than_5_mask = arr < 5

arr[circle_mask & less_than_5_mask] = 1

In this case, & is the bitwise, elementwise boolean AND operator. If you wanted to replace locations that were in the circle and/or less than 5, you could use the | operator instead.

& is a faster eqivalent to np.bitwise_and in this case. Since your arrays are boolean, you can use np.logical_and to the same effect. Note that that is not the same as using the and operator, since the operator will attempt to distill the entire array into a single boolean value, causing an error.

user_29887300
Related