Re: [PLUG] which image transfer format?

[Jeff Abrahamson <jeff@purple.com>]

On Fri, Oct 18, 2002 at 08:13:39AM -0400, Kevin Brosius wrote:
> Jeff Abrahamson wrote:
> > 
> > On Tue, Oct 15, 2002 at 03:41:18PM -0400, Jeff Weisberg wrote:
> > > [2] the real world is not 24 bit color
> > 
> > Actually, neither is your digital camera, unless maybe it's extremely
> > high end. Most digital cameras have eight bit CCD's with color masks
> > over them, so, essentially, one pixel is 8 bit red, the next 8 bit
> > green, the next 8 bit blue, repeat. (In practice, it may use a
> > different color space, and so forth.)
> 
> Umm... In computer graphics, 24 bit color is considered 8 bits for each
> of RGB, right?  What am I missing here?
> 
> Oh, I see, you're referring to the practice of only storing one color
> plane _per_ pixel on digital cameras.  That doesn't really make them 8
> bit color cameras though.  They still have 24 bits of color depth.  Just
> 1/3 the resolution (width x height), right?

Well, the image you get from your camera has eight bit color depth
because of the interpolation. (Try "identify" on one.)

I'm not sure if what you say is equivalent, I haven't studied enough
information theory. But I think essentially that information is lost
in the interpolation, so you can't actually recover a 24 bit image,
even at 1/3 (1/9) resolution.

-- 
 Jeff

 Jeff Abrahamson  <http://www.purple.com/jeff/>

_________________________________________________________________________
Philadelphia Linux Users Group        --       http://www.phillylinux.org
Announcements - http://lists.netisland.net/mailman/listinfo/plug-announce
General Discussion  --   http://lists.netisland.net/mailman/listinfo/plug