Product: Glasses: Color
"seeing" bird colors   (+4)  [vote for, against]
specially tinted glasses to distinguish extra colors

I was just reading a web page that said that birds can see more primary colors than people can.

Instead of three primary colors (red, green and blue), some birds can see four or five primary colors.

So certain ornithologists might be missing out on part of how the birds see the world.

(How could this be important? Maybe some male and female birds might look the same to us, but not to each other because of these colors. Maybe certain fruit are distinct in these colors.)

The glasses would have two different lenses, that to us would look like the same color tint, but that would have different behaviours for different light frequencies, approximating some of the differences perceptible in bird vision.

This is almost the same idea for the glasses that color blind people might wear, only taken up a step.
-- talldave, Oct 18 2003

Color Vision of Birds http://www.users.mi...h/lighting/cvb.html
[talldave, Oct 04 2004]

Color - Wikipedia http://en.wikipedia.org/wiki/Color
[talldave, Oct 04 2004]

Berlin and Kay "basic colors" theory http://www.icsi.ber....edu/wcs/study.html
Actually, this link is a bit critical of the theory. [lawpoop, Oct 04 2004]

Alien View Screen http://www.halfbake...ien_20View_20Screen
Should work for birds, too. [DrCurry, Oct 04 2004]

Photochromic Paintball Lens http://www.unlikely...ews/ravennuvis1.htm
Brings out greens and reds. [Letsbuildafort, Oct 04 2004]

Metamerism http://en2.wikipedia.org/wiki/Metamerism
another article discussing the difference between spectra and hue [talldave, Oct 04 2004]

Article About Women Who See 4 Colors http://www.post-gaz...6256/721190-114.stm
Article About Women Who See 4 Colors [drememynd, Aug 13 2007]

This was one of my first ideas (glasses which extend human vision into invisible wavelengths). I took mine down because I decided the technology I was using was a little too magical. You don't specify a technology at all.

This is possible today with the appropriate electronics, but I don't think it would fit on the bridge of your nose.
-- phoenix, Oct 19 2003


Isn't this similar to what night vision goggles do? I'm no expert on military hardware or vision technologies, but I would imagine that it wouldn't be too difficult to adapt the technology to read IR/UV instead of heat? The main problem I can see is that the resultant image will still be limited to our visible range, and so although we may be able to see things hitherto unseen, they will still be in the same old colours. Short of bypassing the eyeball and tapping directly into the visual cortex I can't envisage a way around that. Mr LaForge?
-- egbert, Oct 19 2003


My thought was to either translate out-of-human-spectrum colors into colors we can see or modify some other aspect of the visual image. Objects reflecting a lot of UV show up brighter, for example. Objects radiating a lot of IR might cast an aura, for another.
-- phoenix, Oct 19 2003


There seems to be a little confusion about the idea....

Consider yellow light: It could be made from the yellow part of the spectrum, or a combination of red light and green light from the spectrum. These two yellows can be visually identical (if the balance of red and green is just right), but a color filter can make them distinguishable.

I'm not talking about using light from outside the normal visual range, I'm talking about adding to the ability to distingiush certain colors that look the same, but have different spectral compositions.

These glasses could also be useful for intro science courses to help clear up some misconceptions about color perception.

The technology involved would be nothing more high tech than knowing which elements, when added to glass, change the transmission of light in what ways. I'm fairly sure that glass makers or material scientists or both have tabulated this already.

Then the next step is to find combinations that are the same in color space (an abstract space with red, green, and blue axis) but with different spectral properties.

The importance of the lenses being the same "color" is that the distinction between what the left eye and the right eye (different lenses for each eye) sees should be as close to the distinction between different ways of making the same color as possible.

In fact, if possible, it would be best if they looked gray, or only very lightly one color. This would happen if white light passed through the filter would still remain white, but have a color balanced amount of different parts of the spectrum removed.

There might also be military applications. Camouflage might look like a convincing match of the background in ordinary colors, but might stand out if other distinctions can be made.
-- talldave, Oct 19 2003


WOW, more than 3 primary colors?! that will entirely wreck the color theory! +
-- synergy~, Oct 19 2003


// I'm talking about adding to the ability to distingiush certain colors that look the same, but have different spectral compositions. //

That doesn't make any sense. Yellow is yellow. How could you know how it got to be yellow? A red or green filter would have the same effect on either source you mention. The spectral composition is the same once the hue is achieved. I don't see how it would be possible to reverse engineer a color.

You're tossing around a term - "different spectral properties" - as though it means something, but it doesn't. Look at these two shades of red. They look the same, don't they? Only they have "different spectral properties". What? It's meaningless.
-- waugsqueke, Oct 19 2003


[Waugsqekeu], are you sure about that?

Split up white light in a prism and from this rainbow you get a bunch of colors including:

red light, wavelenth=690 nm

yellow light, wavelength=590 nm

green light, wavelenth=520 nm

(These numbers are a little rough, but pretty close.)

Combining the red and green light from a rainbow makes yellow light, but shine this light through a prism and it still splits them into red and green light.

Yellow light from a rainbow won't be split up by a prism. If you pass it through a prism, you will see yellow light.

This is what I mean by "spectral properties", if you look at the brightness of light at each wavelength.

So, for instance, if one lens had partial absoption at the red, green and blue parts of the spectrum, while the other had it at yellow and blue they could appear to be the same color, but when you looked through them yellow made from spectral yellow light (the yellow you see in a rainbow) would look different than yellow made from red and green light.

And [Waugsqekeu], if you don't believe me, check out the link, to "color" on the Wikipedia website.
-- talldave, Oct 19 2003


Nice idea, [Rods Tiger]! Write it up on its own page and I'll give it a positive vote.
-- talldave, Oct 19 2003


I think the only way we could implement this is with false color. The extra primary colors birds see are because their brains are also wired to percieve them. Our brains are wired to percieve only certain primary colors. We just can't see "bee's purple" (ultraviolet found on flowers). The best we can do is use a ultraviolet film to capture it, and it gives us back a black-and-white movie with bee's purple being bright white.

Berlin and Kay did a study and found that there are 11 "basic colors" that people can percieve. See link.
-- lawpoop, Oct 19 2003


"I'm talking about adding to the ability to distingiush certain colors that look the same, but have different spectral compositions."
"Combining the red and green light from a rainbow makes yellow light, but shine this light through a prism and it still splits them into red and green light... Yellow light from a rainbow won't be split up by a prism. If you pass it through a prism, you will see yellow light."

I don't buy either of these assertions.
-- phoenix, Oct 19 2003


I buy both of those assertions. An analogy with sound - if a low-pitch note and a high-pitch note (aka red and green) play at the same time, the result is very different from a medium-pitch note, though to the human eye they're both the same (yellow). Relative to the visible spectrum, the eye has a fairly wide-band response to each of the 3 primary colours.

In fact, the eye doesn't filter for reds, greens and blues; instead it detects yellowish greens, greens and blues. It/the brain then calculate red by subtracting the green from the yellowish green (the yellowish-green response is quite wide, covering red to green, but peaking at yellowish-green).

talldave, I'm almost sufficiently interested in your idea to buy some filters and have a play.
-- benjamin, Oct 20 2003


That's a poor analogy, benjamin. A low note and a high note played together sound like two notes, not like a single note in the middle.

Thanks phoe... I needed some back-up on this. I'd been researching talldave's assertions and I can't find anything to back it up.
-- waugsqueke, Oct 20 2003


One thing I've always wondered: why does red look so much like purple if they are on opposite ends of the color spectrum? Is it the subtraction of the yellow / green that makes it so?

What wavelength is burgundy? Very long or very short?
-- RayfordSteele, Oct 20 2003


I need special glasses so I can see how this idea would work.

If birds get five and we have three, how are we supposed to see five even if you show them to us?
-- k_sra, Oct 20 2003


Yeah it's about perception, not about the components of the thing being perceived.
-- phoenix, Oct 20 2003


If you don't buy the premise, check out the Wikipedia link, look under the section "Reproduction of Color".

I don't think this will let people actually "see" bird colors, no more than the lenses colorblind people use let them "see" those colors. It will, let people see that there is a difference, by virtue of the difference between what the left and right eye see.
-- talldave, Oct 20 2003


I agree with [talldave] and [benjamin]. The only reason that sound is a bad analogy is that the human ear can directly sense a very large number of frequencies individually, but the eye can only sense 3 color ranges and interpolates to sense the others. In addition humans can hear pretty well from 50Hz - 15000Hz: over 8 octaves, but the visible range of light 405 - 790 THz is just less than one "octave".

There is one slight problem with this idea however. You cannot create two filters that will both appear to only let yellow light through yet one only lets pure yellow, and one lets yellow composed of red and green, at least not using a simple glass or plasic lense. The problem is that the red and green filter will also let red and green through. What you would need is a lense that only lets green through when it was accompanied by a matching amount of red. I can't imagine how that would be possible.

Of course if you had a filter that only let yellow through, and a filter than only blocked yellow, switching back and forth could still yield some insight on what yellow you see is pure yellow, but the two lenses wouldn't look like they were the same tint as you said in the original idea description.
-- scad mientist, Oct 20 2003


[RayfordSteele] There is no wavelength of the color burgundy. Burgundy can only be composed of more than one frequency, otherwise we would see it in a rainbow. And as this whole discussion suggests, there would be more than one way to make burgundy that looks identical, but could be distinguished by using various filters.
-- scad mientist, Oct 20 2003


Make the title "seeing" bird colors instead of seeing "bird colors" and I'll give you my vote.
-- k_sra, Oct 20 2003


I think that you're probably going to need more than two lenses. If you made a strip of coloured filters, each of which corresponded to one of the primary colours for a species of bird, an experienced operator could probably spot things that appeared only using one filter. It would be time-consuming, though.

Being a fan of gadgets, I'd be more tempted to go for a video camera approach, recording each bird-primary colour as a separate channel and combining some or all into a false-colour image later.

To the nay-sayers: Enter a room. I have placed two buckets on the floor so that their sides touch. One bucket is labeled 'red', one bucket 'green'. Both buckets are half-full with water. Did I fill them by pouring some water into the red bucket and some into the green, or did I pour a larger quantity of water on the point where the buckets touched?
-- st3f, Oct 20 2003


3f, thanks for your analogy. The answer is - it's impossible to tell, which was my assertation to begin with. Yellow is yellow - how it got to be yellow, you cannot determine after the fact.

talldave, this is twice now you've referred people to the Wikipedia page. If there is actually text on that page that supports your presumptions, could you point it out? I've been over it three times now and have not found a) anything to support your statements, and b) anything I didn't already know.
-- waugsqueke, Oct 20 2003


[Waugsqekeu], here is a quotation of the section of the encyclopedia page that I was talking about:

" Reproduction of color

Two different light spectra which have the same effect on the three color receptors will be perceived as the same color. This is exemplified by the color cyan: cyan is a pure spectral color whose wavelength is located just between the responsitivity peaks of the "green" and "blue" cones. A cyan color experience can thus also be generated by an equal mixture of those two peak wavelengths, as long as these don't stimulate the red receptor. The human eye (as opposed to the bird's eye or the spectroscopist) then won't be able to tell the difference between pure spectral cyan and green-blue mixed cyan.

In the same way, most human color perceptions can be generated by a mixture of three colors called primaries. This is used to reproduce color scenes in photogaphy, printing, television, and other media. "

Their example is almost the same as mine, only they use

blue+green=>cyan

cyan=>cyan

where I used

red+green=>yellow

yellow=>yellow

But the idea is the same.

I usually try to avoid appeals to authority in matters I think I understand, but you seem so convinced of your position that I think this might be persuasive.
-- talldave, Oct 20 2003


Yeah, room, walk, bucket, yadda, yadda, yadda.
I think some of you are confusing additive colors and subtractive colors, but I could be wrong.

[talldavu] From your Wikipedia link:
"...one may formally define a color to be the class of all those spectra which give rise to the same color sensation."
yellow=yellow
and a rose by any other name...

I see where you're going with this, but I wonder what the human brain would do if you shined a red light in one eye and a green light in the other. Nothing useful, I'd imagine. Again, I could be wrong.
-- phoenix, Oct 20 2003


Just so I'm taking both sides, I'd agree with [phoenix] here. Consider ye old red/green 3D movies. Each eye is getting a completely differently coloured image, yet the result looks black+white (at least to my brain).

Probably a camera based solution that used flickering/cross-hatching to identify 'bird colours' would provide a better representation.
-- benjamin, Oct 20 2003


I hate to pull the discussion away from the idea but I guess we have to have a consensus on the science to really start on that. Back to the buckets...

You can't distiguish between the two methods of filling the buckets because the buckets represent human vision. We have colour receptors in our eyes tuned to red, blue and green wavelengths. We have no receptors that are tuned to yellow light.

Since each of the three colour receptors have wide, overlapping ranges of wavelengths they can detect, this is not a problem. When 590nm wavelength (yellow) light is shone into the eye, it interracts with both the red and the green receptors and we see yellow. Now, if we shine a little red and a little green light into the eye, this also interacts with the red and the green receptors and we also see yellow. We have no way of distinguishing the tickling of red and green receptors that comes from a single wavelength yellow, and the tickling of the red and green receptors that comes from mixing red and green light.

If you're still with me, I've got a bit of pedantry that will blow the top of your head off: Red and green does not make yellow. Red and green makes a colour that humans perceive as yellow.

If you created a creature that had four colour receptors: Red, green and blue tuned as a human, but added yellow, they would perceive yellow (single wavelength true yellow) as yellow, but would perceive a mix of red and green as a new and different colour. We have no name for this colour since we are unable to differentiate it from yellow. Red, green and blue are *only* considered primary colours because we live in a world populated by humans.

Put a third bucket between the other two. Label it 'Yellow'. Now, can you now tell the difference between a mixed yellow and a pure yellow?
-- st3f, Oct 21 2003


Somewhere back there someone told me blue, red, and yellow were the three primary colors. Since when is green a primary? I know it's on my computer screen...

And thank you, [st3f], that's pretty much my thoughts on this. You can lead a man to color, but you can't make him perceive it.
-- k_sra, Oct 21 2003


Red, Green and Blue are the primary additive colours for when you're dealing with light.

Cyan, Magenta and Yellow are the primary subtractive colours used for paint and inks. Artists generally call these Blue, Red and Yellow, possibly out of a dislike of words that are difficult to spell.

Both painters and printers will generally add black to their standard pallette, as it's difficult to get a good black from anything but the purest primary subtractive colours. They will then argue until closing time as to whether it is actually a colour.
-- st3f, Oct 21 2003


//Make the title "seeing" bird colors instead of seeing "bird colors" and I'll give you my vote//

Make it "seeing" bird colours and you get my (+)
-- silverstormer, Oct 21 2003


...but I take mine away.
-- k_sra, Oct 21 2003


Thanks guys, I feel like I understand the human perception of colour so much better now. And it’s always nice to see a good scientific explanation defeating a "you are wrong" statement with no argument to back it up. I have a further question though. If you were born with a mutation that afforded you a set of extra colour receptors tuned to (for example) the pure yellow wavelength, how might you be discovered? As soon as you were able to articulate the words about colour and hence communicate using them, you would agree with those around you about what was yellow and what wasn't (because you would learn what was yellow and what was not by looking at things and being told if they were yellow or not). But you would (presumably) be able to detect two different flavours of yellow (pure vs composite). Specifically, my question is, would you be able to mix up two different 'yellow' paints to represent your two yellow colours - even though I couldn't see the difference? I could analyse the composition of the paints and at least agree with you that they were different.
-- dobtabulous, Oct 21 2003


our colour perception is provided by the cone cells lining the retina. these come in three varieties containing pigments that are respectively tuned to red, green and blue light. there are genetic mutations that can shift the frequency response of the pigment molecules for a particular variety of cone or reduce their sensitivity (colour blindness) but normal eyes will react in the same way to colour because we all have the same ratio of the same three pigments in our cones.
-- po, Oct 21 2003


I wonder if each of us experience water as varying degrees of wetness :)
-- po, Oct 21 2003


I've thought of another application:

With certain dyes (they would have to age the same) this basic idea could be applied as a counterfit countermeasure.

A normally invisible pattern made up of alternations of two versions of the same color could show up under this filter, but would not be reproducable using scanners or color copier.

This wouldn't stop the professional counterfitters, but would make all casual counterfitting easy to detect.

A bill could be passed quickly under a color filter, and instantly verified to be legit.
-- talldave, Oct 21 2003


bird's bills?
-- po, Oct 21 2003


[Phoenix] the answer is exactly in the quote you presented:

"...one may formally define a color to be the class of all those spectra which give rise to the same color sensation."

This means that there are a number of different spectra that correspond to the same color, that's what the phrase "class of spectra" is refering to.

So two lights with colors that looks exactly the same can be physically, spectrally different.

The main idea is to exploit this difference with some simple optics so that just by looking around someone could "see" if there was something interesting that birds could see, but that we would normally not notice.
-- talldave, Oct 21 2003


I've given the method some thought and have a preliminary way to design these lenses.

I don't remember exactly, but an optically active media (one with molecules with a particular chirality) rotates the polarization of light, the amount depending on the wavelength and the thinkness of the media. If a media like this was placed between polarizers, then only certain frequencies of light would be able to be transmitted.

Now if we could then take advantage of a numerical cooincidence (if it exists) of two different configurations that let through different sets of frequencies of light and those sets had the same hue, then it may be possible to quickly distinguish between, say, monochromatic (one frequency of light) cyan and cyan composed of monochromatic blue and green (each a single frequency).

n.b. Chiral molecules are ones with distinct left handed and right handed versions.
-- talldave, Nov 14 2003


Good idea. (+)
-- galukalock, Jun 04 2006


hey, I read about this today - there's something that has 10 cones. I'll get back later...
-- po, Jun 04 2006


Watched Mr Attenborough's splendid 'Life of Birds' yesterday. You need to extend your visual accuity into the UltraViolet range if you really want to see birds in their full finery.
-- DrBob, Jun 05 2006


[+] for the idea, and the discussion.

After reading the discussion, I was all set to post a link to an article about tetrachromats (women who see four primary colours), only to look up and find that [drememynd] just did so already.

Nice one, [drememynd]. That link is well worth a read - interesting, surprising, and totally relevant to this discussion.
-- imaginality, Aug 13 2007


Thank you [imaginality], It's a subject that has been fascinating to me for years. I was excited to find a discussion where the topic would be relevant.
-- drememynd, Aug 14 2007



random, halfbakery