orchard_bouman

Python implementation of unweighted Orchard Bouman clustering

Requirements:

• NumPy
  Developed in Python 3.10, but anything over Python 3.7 should work

Installation:

pip install orchard-bouman

Usage:

Once your image is created as a NumPy array, simply use:

• ob = OrchardBouman(image, k)
  where image is the image you want clustered in RGB format and k
  is the number of times to split nodes. The final number of nodes = 2^k.
• This will return the split node objects within a list accessible by ob.nodes
• A clustered color image can be generated by ob.construct_image()

Considerations:

This package does not handle image reading and saving to keep it versatile.
Because all images are treated as a NumPy array, it is recommended to
read in images using OpenCV, as they are natively NumPy arrays.