Page 1 of 1

### The Reverse of how colors work: how would it look?

Posted: Sat Mar 18, 2017 5:42 pm UTC
It's a simple question:

Upon looking at this drawing, I began to wonder what the object in Our universe (colored red when it absorbs all wavelengths of visible light except for red), would look like instead if it absorbed ONLY red and reflected all other colors, like an opposite. Moreover, what would an object look like if it's color was a solid "all colors except for red"? Does it just blend?

I understand slightly the answer, but I can't really put it into words and I have to explain it to a Younger one.

### Re: The Reverse of how colors work: how would it look?

Posted: Sat Mar 18, 2017 5:58 pm UTC
If it absorbed red and reflected all other colors, it would look cyan.

### Re: The Reverse of how colors work: how would it look?

Posted: Sat Mar 18, 2017 5:58 pm UTC
In the setup you have above, assuming specific red rays, green rays and DVDs blue rays, a red-absorbing object would reflect green and blue, which would result it in being seen as cyan.

But that assumes a rather specific lighting. Natural lighting is a spread of frequencies, with gaps that you don't notice unless you're looking for them. i.e. an object that was 'red' in that it absorbed only the red light at (say) 680nm, though, does it not absorb light at 640nm (still red)? 660nm (a closer red)? 670nm? 679nm? Given a white light source, the narrower the absorption, the more red-but-not-exactly-as-red light there is that still reflects to work with the reflected non-reds to appear "white" (or nearly so) to the Mark I Eyeball.

See https://en.wikipedia.org/wiki/RGB_color ... n_and_blue for the whole "why RGB"/vision thing...

### Re: The Reverse of how colors work: how would it look?

Posted: Mon Mar 20, 2017 2:24 am UTC
The "red object" in your drawing looks magenta to me.

### Re: The Reverse of how colors work: how would it look?

Posted: Tue Mar 28, 2017 6:09 pm UTC
Note that this "absorbs one color and reflects everything else" is the way that *most* objects in the real world operate. Reflecting only one color is relatively rare - that's a lot of energy to absorb, and molecules usually only have a small band of wavelengths they can absorb, so you need either a very good mix of molecules covering the absorption spectrum, or some very special molecules that have extra-wide absorption bands (like that hyper-black coating that's making the news rounds again).

For example, apples are red because they absorb greenish light and reflect everything else.

### Re: The Reverse of how colors work: how would it look?

Posted: Thu Mar 30, 2017 3:52 pm UTC
If you reversed absorption and reflection, things would generally look like their colors in a photographic negative.

(A whole scene wouldn't look like that, though, because changing the surface properties of everything wouldn't result in different shadow patterns, but negatives have bright shadows and dark light sources.)

It's also complicated by the fact that photographic negatives look the way they do because photographic technology is designed with human color vision in mind. Two things that look the same color in real life should also have the same color in a photograph and in the negative of that photograph. But the actual reflection and absorption spectra of the two things might be quite different, and they only coincidentally look identical to human trichromatic eyes. Inverting those spectra might bring out some of the differences we couldn't have noticed before.

(This is related to the fact that reflections of RGB monitor or TV or fluorescent light don't look naturally colored, even if the monitor itself looks identical to a full-spectrum light.)

### Re: The Reverse of how colors work: how would it look?

Posted: Thu Mar 30, 2017 5:40 pm UTC
Here are a pair of images that approximately illustrate what you're asking about. The colours of one image are the RGB complement of the other image, so the hues of corresponding pixels are 180 degrees out of phase, but they have the same saturation and value (brightness). Well, sort of for the reasons Gmal mentioned.

Those images can also be seen on this Stack Overflow page, along with the Python code I wrote to perform the colour transformation.