rolf@warwick.UUCP (03/10/87)
With the recent discussion about how to map colour images to black-and-white,
I thought I'd ask if anyone could explain to me how the human brain
perceives colour.
One thing that always puzzled me at school was being told, for example, that
red + green light "gives you yellow", where "yellow" is also what you see
at a particular position in the spectrum (when you shine white light
through a prism). As far as I understand it, these two "yellows" are different
spectroscopically , yet the eye perceives them to be the same colour.
I think I was confused because the distinction between what a colour *is*
and how it *appears* to the eye was never made very clear.
Obviously the effect of being able to combine three primary colours to give
a complete (?) spectrum is vital for things like colour television (reducing
the information needed to describe a colour from potentially infinite to about
3x8 bits). Few people seem to be willing to admit that this is only an illusion
though, and is highly dependent on the properties of the human eye and brain.
Is it not conceivable that someone with a certain type of colour defect might
recognise a yellow banana in sunlight, but when shown a picture of one on TV
say "I can only see superimposed red and green images" ? Certainly machines
aren't fooled the way we are, for example, one can't do a spectral analysis of
the television image of a sodium street light and expect to get anything like
the original.
Sorry if I have rambled on a bit, but I would be very grateful for any
help out of my present confused state.
--
Rolf
Dept. of Computer Science, Tel: +44 203 523523 ext 2485
Warwick University, JANET: rolf@uk.ac.warwick.ubu
Coventry, UUCP: {seismo,mcvax}!ukc!warwick!rolf
CV4 7AL
England. "Three pints? At lunchtime?!"jsgray@watrose.UUCP (03/12/87)
When you perceive the same colour from a monochromatic yellow light and from an image on your television set, you are experiencing the phenomenon known as metamerism. The eye has three different types of colour detecting cones. These cones each respond to a wide spectrum of colour, with maximum response situated at a certain red, green, and blue wavelength. When you irradiate these cones with a monochromatic yellow light, the red and green cones are stimulated, but the blue cones (whose spectral response is small or none for that yellow light) are not. You perceive "yellow". When you simularly irradiate these same cones simultaneously with the red and green light from the phosphors on your television screen, the red and green cones are stimulated, but the blue cones are not. You perceive the same "yellow". Thus it is quite possible to perceive the same colour from very different spectral distributions (which are called metamers). [I hope the colour experts here at UW will forgive my transgressions.] Jan Gray jsgray@watrose University of Waterloo (519) 885-1211 x3870
chassin@rpics.UUCP (03/12/87)
In article <505@ubu.warwick.UUCP>, rolf@warwick.UUCP (Rolf Howarth) writes: > > With the recent discussion about how to map colour images to black-and-white, > I thought I'd ask if anyone could explain to me how the human brain > perceives colour. > > One thing that always puzzled me at school was being told, for example, that > red + green light "gives you yellow", where "yellow" is also what you see > at a particular position in the spectrum (when you shine white light > through a prism). As far as I understand it, these two "yellows" are different > spectroscopically , yet the eye perceives them to be the same colour. > I think I was confused because the distinction between what a colour *is* > and how it *appears* to the eye was never made very clear. > I not very expert on physiology but as far as I understand it color is an interaction between a mass of photons, the cones in the retina, and the coding sent to the brain. There are I believe two types of color sensing cells in the retina: red-green and yellow-blue. I don't know how the cones send their info to the brain via the optic nerve. I was told however the the cones react only to their type of light. A red-green will swing to the red side in response to a red low frequency photon and swing over to the green side as the photons get more energetic (greener color). A yellow blue does the same thing but at a higher frequencies. There is an overlap which I believe accounts for the strongly favored green region. This makes sense since this is the peak output frequency of the sun. I also was told that the response curve actually matches the radiation emission curve for the temperature of the sun. Sorry I can't give you sources, or more detail, I don't know much more... I'm sure other people will be able to give more detail and/or likely corrections. I do want to comment that this is an interesting subject to me in terms of human perception, and video displays. Their is no doubt in my mind that a video display offers an incomplete picture, so to speak. To be more complete a display would have to either be capable of generating all frequency photons rather than three specific. The other option would be to generate to ranges of light that can affect the retina in a way equivalent to the way 'natural' light does. Either of the ways would be transparent, giving the real colors. _____________________ David P. Chassin Rensselaer Polytechnic Institute | School of Architecture __+__ Troy, NY 12181 / _ \ USA | | | | /=======/ = \=======\ (518) 266-6461 | _ | _ | _ | | | | | | | | | | | chassin@csv.rpi.edu | = | | | | = | =======================================================================
ksbooth@watcgl.UUCP (03/12/87)
The SIGGRAPH tutorial notes from 1984 or 1985 on colour give a good overview of human colour vision. There some introductory material on the topic in the collection papers that I edited for IEEE (the title is "Tutorial: Computer Graphics" available from IEEE Computer Society). This is not a trivial topic. Hopefully we will be spared too many novice replies to the previous posting requestng this information.
gary@uoregon.UUCP (03/14/87)
In article <505@ubu.warwick.UUCP> rolf@ubu.UUCP (Rolf Howarth) writes: > >I thought I'd ask if anyone could explain to me how the human brain >perceives colour. > The retina of the human eye contains receptors with spectral sensitivities that peak in the short, medium, and long wavelength regions of the visible spectrum. Any spectral energy distribution has therefore been reduced to three pieces of information by the time it passes this initial stage of the visual system. If two spectral energy distributions produce the same three signals then the visual system will not be able to tell them apart. This is the principle upon which all color reproduction techniques are based. Any visual system with spectral sensitivities different from those used to perform the color calculations will not be satisfied with the color reproduction (although two separate images would not be seen as you suggest). All animals and some color defective humans fall into this category. However people whose color deficiency stems only from the lack of one of the spectral sensitivity functions will still find the color reproduction adequate. These indviduals are known as dichromats. I have become convinced that most of the confusion which surrounds this topic is caused by the use of perceptual terms (red, yellow, green, etc.) to label physical quantities (wavelength, spectral sensitivity). Hue is a term of identification and its use in this context leads to the misconception that this is a color naming task when it is really the much simpler problem of color matching that is involved. I pass along the following quote which helped to "enlighten" me. And if at any time I speak of Light and Rays as coloured or endued with Colours, I would be understood to speak not philosophically and properly, but grossly, and accordingly to such Conceptions as vulgar People in seeing all these Experiments would be apt to frame. For the Rays to speak properly are not coloured. Sir Isaac Newton Opticks, Book One, Part II
gwyn@brl-smoke.ARPA (Doug Gwyn ) (03/14/87)
There are many aspects to color and its perception. One important thing to be aware of is that there are colors that are not in the "spectrum". Another is that the eye does not contain simply a continuous frequency-and-intensity-measurer, but a complex collection of detectors the composite output of which is combined, along with other mental processes (including some psychological ones) to determine what one "sees". It isn't possible to come up with a positive-definite metric for "color space". I read some work done long ago by E. Schr"odinger and perhaps O. Veblen on this. I hope some specialists in this field will post a few good references describing color vision.