colour_space.md

Background

One way to encode colour images is with:

• a rectangular grid of pixels (the more pixels, the more detail)
• 'RGB' values for each pixel, registering the relative strength of red (R), green (G), and blue (B).

Task

1. Type: Explore
   • Import an image library and cat picture e.g.,:
     • Import the scikit-image library (import skimage as ski)
     • Open their stock image of a cat (cat = ski.data.chelsea(), ski.io.imshow(cat), ski.io.show())
   • Explore the shape of the image (cat.shape)
   • Retrieve individual pixels and individual colour (RGB) profiles of each pixel
2. Type: Implement
   • Write a function to modify the colour profile by weighting the R, G, B components of each pixel.

Reference implementations are in the notebook (below), e.g. modify_colour().