fastest way to find the rgb pixel color count of image

by

You need to use Numpy, or OpenCV, for fast image processing in Python. I made a 9-colour version of Paddington:

enter image description here

from PIL import Image
import numpy as np

# Open Paddington and make sure he is RGB - not palette
im = Image.open('paddington.png').convert('RGB')

# Make into Numpy array
na = np.array(im)

# Arrange all pixels into a tall column of 3 RGB values and find unique rows (colours)
colours, counts = np.unique(na.reshape(-1,3), axis=0, return_counts=1)

print(colours)
print(counts)

Results

[[ 14  48  84]
 [ 19  21  30]
 [ 33 108 163]
 [ 33 152 190]
 [ 72  58  58]
 [ 96 154 210]
 [180  89  64]
 [205 210 200]
 [208 151  99]]

[20389 40269 12820  1488 17185 25371 17050 16396  9032]

That means there are 20,389 pixels of RGB(14,48,84), and so on.

That takes 125ms on my Mac for a 400×400 image, which will give you 8 fps, so you better have at least 4 CPU cores and use all of them to get 25+ fps.

Update

I think you can actually go significantly faster than this. If you take the dot-product of each of the pixels with [1,256,65536], you will get a single 24-bit number for each pixel, rather than 3 8-bit numbers. It is then a lot faster to find the unique values. That looks like this:

# Open Paddington and make sure he is RGB - not palette
im = Image.open('paddington.png').convert('RGB')

# Make into Numpy array
na = np.array(im)

# Make a single 24-bit number for each pixel
f = np.dot(na.astype(np.uint32),[1,256,65536]) 

nColours = len(np.unique(f))     # prints 9

That takes 4ms rather than 125ms on my Mac 🙂

Keywords: Python, Numpy, PIL/Pillow, image processing, count unique colours, count colors.

More Related Contents:

  1. Pillow installed, but getting “no module named pillow”
  2. How to merge a transparent png image with another image using PIL
  3. scipy.misc module has no attribute imread?
  4. Python Image Library fails with message “decoder JPEG not available” – PIL
  5. Get pixel’s RGB using PIL
  6. Image library for Python 3
  7. Change the colors within certain range to another color using OpenCV
  8. How to convert a NumPy array to PIL image applying matplotlib colormap
  9. The problem with installing PIL using virtualenv or buildout
  10. How do I convert a PIL Image into a NumPy array?
  11. Changing image hue with Python PIL
  12. How do you merge images into a canvas using PIL/Pillow?
  13. Center-/middle-align text with PIL?
  14. Python: PIL replace a single RGBA color
  15. TypeError: __array__() takes 1 positional argument but 2 were given (Image classification Keras)
  16. How do I get the picture size with PIL?
  17. Can you “stream” images to ffmpeg to construct a video, instead of saving them to disk?
  18. How to Split Image Into Multiple Pieces in Python
  19. Convert RGB to black OR white
  20. Is it possible to create encoded base64 URL from Image object?
  21. Can’t install PIL after Mac OS X 10.9
  22. How do you composite an image onto another image with PIL in Python?
  23. How to show PIL images on the screen?
  24. Python: Remove Exif info from images
  25. create an image with border of certain width in python
  26. Change EXIF data on JPEG without altering picture
  27. Python/PIL Resize all images in a folder
  28. Why is the quality of JPEG images produced by PIL so poor?
  29. How to get alpha value of a PNG image with PIL?
  30. Python: Converting GIF frames to PNG

Leave a Comment