Wow! I wonder if we could we potentially infer color from black & white photos of celestial objects, by this technique? Maybe it could be combined with the adaptive optics that attempt to cancel stuff like this?
you can already do similar things with filters, you don't need any exotic biology to do it.
But filters block light, reducing the effectiveness of the telescope. Also, you need to set them up before the photos are taken. I am suggesting a way to process existing grayscale photos, where the operator may have decided that filters weren't worth diminishing the captured light, or the increase in communications bandwidth.