Re: [Phys-l] An Unsolved Color Problem

["clement" <clement@hal-pc.org>]

> Here is excellent web link with optical illusions (including the
> checkerboard mentioned earlier) along with rather technical attempts to
> explain the various phenomena.
>
> http://www.michaelbach.de/ot/

The color problem was actually explored many years ago and a model (Retinex
theory) was created for it by Edwin Land.  He presented the model at an APS
meeting.  The web site above did mention that the processing is done in the
brain, or perhaps the optic nerve for many effects.  Land had come up with a
simple circuit that mimics the perception of color and luminance.

His experiment was very simple.  He showed that a matted array of different
color squares kept their apparent color even when the balance between the
red and green light on a red square on one side was the same as for a green
square on the other side.  This was done by a simple arrangement of red and
green floods.  This also works for black, white and grey.

The Howard Hughes Medical Institute http://www.hhmi.org/senses/ has a good
online book exploring the senses, including the color/brightness constancy
of materials.  They go into details and mention the Retinex theory, but it
is sufficiently non technical that a non-biologist can read it.  It could
even be read by students in physics classes.

" "Color constancy is the most important property of the color system,"
declares neurobiologist Semir Zeki of University College, London. Color
would be a poor way of labeling objects if the perceived colors kept
shifting under different conditions, he points out. But the eye is not a
camera. Instead, the eye-brain pathway constitutes a kind of computer-vastly
more complex and powerful than any that human engineers have built-designed
to construct a stable visual representation of reality."

Incidentally it is also possible to reproduce full color with only 2 colors.
But this requires special conditions such as a fully darkened room.

Of course physics books still pretend that the 3 color theory is exact, but
then art teachers teach that one can get all colors by mixing red, green,
and blue (invented? By Goethe).  But the 3 color theory does work very well
for engineering color reproduction.  So the books should really call it a
model of color reproduction rather than a model of vision.  But of course,
some state standards and texts still mandate the 7 colors of the spectrum,
even though indigo is not really a visually distinct color when it can be
perceived.  But after all if Newton said it, it has to be so.

John M. Clement
Houston, TX