How a salience strategy turns the camera's colour image into the channels it reasons about.

Mirrors tbp/monty/frameworks/models/salience/strategies/vocus2/images.py::ColorSpaceConverter, a Protocol in Monty, so a strategy can name any function with this shape. It is a behaviour here: an RGB image in, a float image out, one channel or several, shifted and scaled so that [0, 1] is the whole range.

Monty's three converters — rgb_to_lab/1, rgb_to_opponent/1 and rgb_to_opponent_codi/1 — are plain functions in the module beside this one and are not instances of the protocol; they match its shape, which is all either needs.