🚀 go-pugleaf

dak@tabaqui.informatik.rwth-aachen.de (David Kastrup) writes:

>The rest agrees pretty much with what I know. However, I think I remember
>that JPEG did not use a RGB model, but some other color model. This is
>because changes in luminosity are more easy spotted by the human eye
>than changes in saturation in hue. The exact color model I do not know,
>but I think I remember that about double the number of information was
>kept for luminosity than for the other two color vectors.

There is a different color model.  Luminance (Y) and two dimensional color
(Cr and Cb).  I'm only guessing this is similar to YIQ in NTSC video.  It
is based on the same chrominance theories, so it should be at least rather
close.  Also, the blocking is a bit different for color to gain more
compression.

I'll be rewriting my article with the new information.
-- 
/************************************************************************\
| Phil Howard,  pdh@netcom.com,  KA9WGN    Spell protection?  "1(911)A1" |
| "It's not broken... it's just functionally challenged" --Phil and Pete |
\************************************************************************/