Is there a function to transform all element of ndarray?

Question

I must transform all elements of an ndarray (a grayscale image). In particular if I have the value 0 the desired value is 0 and the values > 0 become 1. Is there a function of numpy that do this? I can do this by two for but I think that it isn't the best solution.

like this? `myarr[myarr > 0] = 1` – Nullman Jul 22 '19 at 10:28 — Nullman, Jul 22 '19 at 10:28
like this myarr[myarr>0] = 1 – Giuseppe Puglisi Jul 22 '19 at 10:28 — Giuseppe Puglisi, Jul 22 '19 at 10:28

score 1 · Answer 1 · answered Jul 22 '19 at 10:30

1

You should look up fancy indexing and/or boolean indexing in numpy. Effectively what you can do is as such

array[array>0] = 1

This will set all indices > 0 to 1

answered Jul 22 '19 at 10:30

Imtinan Azhar

1,725
10
26

score 1 · Answer 2 · answered Jul 22 '19 at 10:58

if you don't want to work "in-place" you can do:

def to_bw(img, dtype=np.uint8):
    return (img > 0).astype(dtype)

e.g. you use this as:

import numpy as np

to_bw(np.array([0,1,2]))

it will evaluate to:

array([0, 1, 1])

note this sounds pretty similar to: Convert RGB to black OR white

Is there a function to transform all element of ndarray?

2 Answers2