Count unique colours in an image

Question

I'm trying to count the number of (unique) colours in an image. This is what I've got:

from PIL import Image

myImage = "D:\\temp\\tiny_pal.png"

img = Image.open(myImage)
size = img.getcolors(img.size[0]*img.size[1])

if len(size) < 256:

  colours = img.convert('RGB').getcolors() # this converts the mode to RGB

  print str(len(colours)) + " colours"

else:
  print ("Image is too big!")

So far so good

>> 43 colours

...Only it's using PIL and moreover, is limited to a rather a small image size.

I'm aware of this question. However, it seems to have been confusingly marked as duplicate for wanting to find most dominant colour in an image.

Can you count colours, like my example above, in larger images with numpy or something similar?

you want colors as visible colors or you want unique pixel values of RGB? — alexzander, Mar 02 '21 at 19:43
Have a look here...https://stackoverflow.com/a/59671950/2836621 — Mark Setchell, Mar 02 '21 at 19:44
`img.getcolors(img.size[0]*img.size[1])` returns a list of all colors, so `len(size)` is all you need. — Anon Coward, Mar 02 '21 at 19:46
@alexzander an array of RGB colours that I could get the `len()` of would be great. — Ghoul Fool, Mar 02 '21 at 19:56

score 1 · Answer 1 · answered Mar 02 '21 at 19:45

How about using numpy. If you are looking for different RGB colours you can use np.unique() specifying the axis correctly to obtain all different colours. Then you could do len of the array obtained to know how many different colours are in the image.

Count unique colours in an image

1 Answers1