Finding the number of pixels in a numpy array equal to a given color

Question

I have been scouring around and can't seem to find an answer to this question. Say I have a given RGB value, i.e. (255,0,25) or something like that.

I have a ndarray called 'img' of shape (height, width, 3). Now, I want to find the number of pixels in this array that equal my color. I thought doing

(img==(255,0,25)).sum() would work, but even if my image is a comprised only of the color (255,0,25), I will overcount and it seems that this is summing up when r=255, 0, or 25, and when g=255,0, or 25, and when b=255,0, or 25.

I have been searching the numpy documentation, but I can't find a way to compare pixel-wise, and not element-wise. Any ideas?

@QuangHoang I tested on 1D list of pixels but for 2D array of pixels it has to be axis=2 and it gives the same as axis=-1 — furas
– furas, Commented May 19, 2020 at 17:58
Exactly my point. In general, it'd better be axis=-1. That said, an image is more likely a 3-D array. — Quang Hoang
– Quang Hoang, Commented May 19, 2020 at 17:59

furas · Accepted Answer · 2020-05-19 18:04:17Z

5

it compares every value in RGB separatelly so every pixel gives tuple (True, True, True) and you have to convert (True, True, True) to True using .all(axis=...)

For 3D array (y,x,RGB) you have to use .all(axis=2) or more universal .all(axis=-1) as noticed @QuangHoang in comment.

print( (img == (255,0,25)).all(axis=-1).sum() )

import numpy as np

img = np.array([[(255,0,25) for x in range(3)] for x in range(3)])
#print(img)

print( (img == (255,0,25)).all(axis=-1).sum() )  # 9

img[1][1] = (0,0,0)
img[1][2] = (0,0,0)
#print(img)

print( (img == (255,0,25)).all(axis=-1).sum() )  # 7

answered May 19, 2020 at 18:04

furas

149k12 gold badges121 silver badges171 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Finding the number of pixels in a numpy array equal to a given color

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related