P3D Re: luminance & color signals

[Jim Crowell <crowell@xxxxxxxxxxxxxxx>]

At 3:08 PM -0700 4/21/98, Rob wrote:
>>
>> It's based on acuity for black and white being superior to all colours.
>> Colour TV uses the same trick (at least it does here, I know nothing about
>> NTSC) of being a sharp B+W signal overlaid with a softer colour one.
>..
>   Yes. its also true of NTSC.  NTSC is actually very similar to PAL in
>basic principle (though not compatible, but listing all the differences
>here would clearly be going off 3d topic).
>   Both take advantage of the fact that the human eye has more
>B&W-sensing "rod" cells than color-sensing "cone" cells.

No, the rods only contribute to vision under very low light levels.  The
signals you're talking about are combinations of outputs of different types
of cone cells (there are three, usually referred to as L, M, and S, for
long-, medium-, and short-wavelength-sensitive).  The high-resolution
luminance signal you're referring to is made by essentially adding the
outputs of L and M cones; the color signals are constructed by differences
between the outputs of various cone types, but as people have pointed out
it's done in a very spatially coarse way.

-Jim C.

----------------------
Jim Crowell
Caltech Division of Biology
Mail Code 216-76
1200 E. California Blvd.
Pasadena, CA 91125
Tel: (626) 395-8337
Fax: (626) 795-2397
jim@xxxxxxxxxxxxxxx
http://vis.caltech.edu/~jim/Home.html




------------------------------