The "important" part is at the very end:

>As light passes through these windows, the machine processes the colour as data, explains Ostadabbas. Machine learning models then look for patterns in order to better identify the corresponding colours analysed by the device.

>“Instead of breaking it down into its principal red, green and blue components, when a coloured light appears, say, on a detector, instead of just seeking those components, we are using the entire spectral information,” says Kar.

"Ordinary" cameras sensors (CCD/CMOS) use filters (e.g. Bayer, which also aims to account for the human eye's sensitivity to green light) to reconstruct an image made up of the three primary colours (RGB). The camera sensor does not actually detect "millions" of colours - it just detects varying levels of the colours passed through by the filter.

As I understand the article, this is a different approach that does not restrict the recorded spectrum in the same way.