neal@druhi.UUCP (05/27/87)
I'm looking for information on color science and computer graphics. 1) Could someone point me to software which converts between the various color systems? For example, from "dominant wavelength, purity, reflectance" to RGB? How about between the Munsell colors and RGB or XYZ? How about software which runs on, say, a Sun lets you navigate around the different 3-d color spaces? 2) How about equations to provide the xy or RGB values for various black body temperatures (when they say that a "G" star is yellow, I'd like to be more specific....) 3) Where would I find specifications on the gamuts of specific color monitors or tv sets? Are they all about the same, or is there significant variation? 4) How would I calibrate a monitor? What do the "hue", "saturation", etc knobs really do? 5) As I understand it, the color palatte of the IBM CGA standard is really limited to 8 colors with 2 intensities of each. On the other hand, the EGA standard allows for a "64 color palatte". Which 64 colors? 2 bits per primary? 6) What are good reference books for questions like these? Thanks a lot for any help! -Neal McBurnett, ihnp4!druhi!neal
dhesi@bsu-cs.UUCP (Rahul Dhesi) (05/27/87)
Keywords: In article <1937@druhi.ATT.COM> neal@druhi.ATT.COM (Neal D. McBurnett) writes: >5) As I understand it, the color palatte of the IBM CGA standard is really > limited to 8 colors with 2 intensities of each. On the other hand, > the EGA standard allows for a "64 color palatte". Which 64 colors? > 2 bits per primary? This problem of terminology comes up again and again. People, the human eye needs only three distinct hues for full effect. This business of 8 colors, or 64 colors, or 4096 colors, is advertising hype. Your typical printer will tell you he or she uses only three colors, with a little black thrown in. The rest is all a question of INTENSITY. Remember, THREE colors, and a large number (or 16, or 64, or 4096) of INTENSITIES. -- Rahul Dhesi UUCP: {ihnp4,seismo}!{iuvax,pur-ee}!bsu-cs!dhesi
dts@gitpyr.gatech.EDU (Danny Sharpe) (05/28/87)
In article <1937@druhi.ATT.COM> neal@druhi.ATT.COM (Neal D. McBurnett) writes: >I'm looking for information on color science and computer graphics. > >1) Could someone point me to software which converts between the various > color systems? For example, from "dominant wavelength, purity, > reflectance" to RGB? . . . >6) What are good reference books for questions like these? _Color_in_Business,_Science_and_Industry 3rd ed., by Judd and Wyszecki, John Wiley & Sons, Inc., 1975, ISBN 0-471-45212-2. This book isn't about software, but it does give some of the zillions of formulas for specifying color. It's rather expensive, and even though it's a book about color it has no color illustrations :-(. ------------------------------------------------------------------------------ "Mistrust enterprises that require new clothes." Danny Sharpe School of ICS Georgia Insitute of Technology, Atlanta Georgia, 30332 ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!dts
todd@uhccux.UUCP (05/29/87)
In article <725@bsu-cs.UUCP> dhesi@bsu-cs.UUCP (Rahul Dhesi) writes: >This problem of terminology comes up again and again. People, the human eye >needs only three distinct hues for full effect. This business of 8 colors, or >64 colors, or 4096 colors, is advertising hype. Your typical printer will >tell you he or she uses only three colors, with a little black thrown in. The >rest is all a question of INTENSITY. You are wrong. Just some of the things color perceptions depends on are: 1. whether the human eye(s) are "normal" for color vision (i.e., the person is not "color-blind") 2. the other colors and shadows surrounding the color surface or source (in the case of lights) 3. the mixture of wavelengths reaching the eye. Subtractive mixture takes place in paints (where Red, Yellow, and Blue are the "primaries"). Additive mixture takes place when using lights (where Red, Green, and Blue are the "primaries") 4. the amount of light reaching the eye 5. the current state of the visual system. If you are dark adapted and are suddenly flooded with light, all you see is a white glare most of the time. If your opponent cells are fatigued in one direction the complentary color (of whatever is fatigued) is seen (try placing a green lens over one eye for a few mintues and then look around). In short, the color you see depends not only on the actual wavelengths reaching your eye, but also interaction of color after being processed by higher cortical functions, and the current state of various portions of the visual system. Further, the while only three "primaries" are needed to create all the colors we "see," the "normal" human eye is capable of distinguishing a great many "colors" (some estimates approach or exceed one billion). Anyone who wants to find out more about the processing of color by the human visual system should read a good book on the subject. I think Robert Boynton's "Human Color Vision" may be a good place to start....todd -- Todd Ogasawara, U. of Hawaii Computing Center UUCP: {ihnp4,seismo,ucbvax,dcdwest}!sdcsvax!nosc!uhccux!todd ARPA: uhccux!todd@nosc.MIL INTERNET: todd@uhccux.UHCC.HAWAII.EDU
eugene@pioneer.arpa (Eugene Miya N.) (05/29/87)
People have not answered one part of the questions.
Black bodies: you want to look at the Boltzmann and Planck equations.
Check out a physics text on radiative transfer and you will find the
curves for black bodies. I learned about them for weather modelling.
Other people have adequately covered the sensation of the perception of colour.
The black body stuff won't help you for color.
From the Rock of Ages Home for Retired Hackers:
--eugene miya
NASA Ames Research Center
eugene@ames-aurora.ARPA
"You trust the `reply' command with all those different mailers out there?"
"Send mail, avoid follow-ups. If enough, I'll summarize."
{hplabs,hao,ihnp4,decwrl,allegra,tektronix,menlo70}!ames!aurora!eugeneawpaeth@watcgl.UUCP (05/30/87)
The Boltzmann equations define the spectal energy of a perfect black-body
radiator. For use in color, these values can be converted into cie
chromaticity coordinates (they are well tabulated), which in turn can
be mapped into RGB coordinates through a change of basis, if you know the
chromaticity of your monitor. Enclosed are some sample values.
/Alan Paeth
---------------
CIE chromaticity coordinates for a black-body at T in deg K.
lambda(max) is the peak wavelength in nm of the Stefan-Bolzmann curve.
Dominant wavelength is not available for ~6500 degrees Kelvin, as the curve is
nearly flat through the visible portion of the spectrum here. Standard daylight
is approximated by illuminant D65, but does not lie along the locus of points
for blackbody radiators.
Temp-K x' y' lambda(max)
100 .735 .265 695
300 .734 .266 684
500 .721 .279 641.5
1000 .625 .345 606.7
1500 .586 .393 594.8
2000 .526 .413 588.9
2500 .477 .414 585.2
2854 .4476 .4074 583.5
3000 .437 .404 582.9
3500 .405 .391 580.9
4000 .380 .377 578.9
4500 .361 .363 577.2
4800 .351 .356 575.1
5000 .345 .352 572.6
5500 .332 .341 0
6000 .322 .332 0
6500 .313 .324 485.7
7000 .306 .371 483.7
8000 .295 .305 481.5
10000 .281 .288 479.4
24000 .253 .253 477.0
99999 .240 .234 475.7
---------------------------------------
3x3 multiply to convert XYZ (cie chromaticity) into RGB
(1) set X=x', Y=y', Z=(1-x'-y') and then
(2) compute
|R| | 1.73 -.48 -.26 | |X|
| | | | | |
|G| = | -.81 1.65 -.02 | * |Y|
| | | | | |
|B| | 0.08 -.17 1.28 | |Z|
where these RGB are those defined the the NTSC which standardizes
broadcast colors for North America television transmission (this is the
inverse matrix of the definition, rounded to three significant places).
/Alan Paethchilds@cadnetix.UUCP.UUCP (06/05/87)
In article <725@bsu-cs.UUCP>, dhesi@bsu-cs.UUCP (Rahul Dhesi) writes: >In article <1937@druhi.ATT.COM> neal@druhi.ATT.COM (Neal D. McBurnett) writes: >>5) As I understand it, the color palatte of the IBM CGA standard is really >> limited to 8 colors with 2 intensities of each. >This business of 8 colors, or 64 colors, or 4096 colors, is advertising hype. >The rest is all a question of INTENSITY. >Remember, THREE colors, and a large number (or 16, or 64, or 4096) of >INTENSITIES. >Rahul Dhesi UUCP: {ihnp4,seismo}!{iuvax,pur-ee}!bsu-cs!dhesi Just what I want Red, Green, Blue; dark Red, dark Green, dark Blue; light Red, light Green, light Blue; etc... Don't you really mean THREE colors, each at its own intensity? The IBM gives you eight different colors, plus a darker or lighter version of the eight, depending on how you look at it. David Childs - Cadnetix Corp. childs@cadnetix.UUCP Disclaimer - I in no way speak for Cadnetix since Cadnetix machines have only eight colors. (All at the same intensity. Except black.)