You are here: DAD-opto.de --> home/english --> index --> sample texts --> Color Essay --> Para. 2.3


Essay on Color, Para. 2, reality in physics (continued)

2.3 Colored Light
-----------------

When even the emission of a 3-band fluorescent lamp can be seen as "white", then what for havens' sake can be colored?

Well, the trick is to stimulate the three cone types (fig.1.2-c in para.1.2) to different amounts. The most striking color effects occur when only one of the cone types is stimulated or when two are stimulated and one is dropped.

This is used to make filters which sift the fundamental colors out of white light. They represent highly saturated colors (though they do not produce monochromaticity).

    fig.2.3: colorful spectral characteristics (188 kByte)

The following table shows which type of cone is mainly stimulated by the light from which filter:

filter cone type
R L
G M
B S
C S & M
M S & L
Y M & L


In "real life", colors usually are not as striking. Their spectral characteristics is not "switching" between 0% and 100%, but there are smooth transitions from a bit depressed levels to a bit elevated levels. Making the colors appear more brownish and greyish (just like most colors in nature).

This way, many natural colors are what we would call "white" if it was an illuminant for an entire scene.
"Spatial resolution" is the key to solve this contradiction. Differently colored surfaces in one scene make us recognize these differences and compensate for any (natural) hue of illuminant.

And it makes clear why every measurement of surface color coordinates MUST also report the type of illuminant used.




Link List and Literature
------------------------


Subject used in source
filter set curves fig.2.3 Edmund Industrie Optik GmbH, "2002 Optics and Optical Instruments Catalog" p.76 (modified)


Continued: 2.4 Some Spectra   Contents of entire essay   Contents of entire web site

Last modified March 11th, 2003; 23:14