Re: Color Accurate Camera Work
Re: Color Accurate Camera Work
- Subject: Re: Color Accurate Camera Work
- From: Robin Myers <email@hidden>
- Date: Tue, 15 May 2001 07:19:48 -0700
- Organization: Robin Myers Imaging
>
> The problem with hue has to be color corrected. It is impossible to know
>
> how the pigments will react to light. Even though we can put a Q60 in
>
> front of the camera and shoot it, it's only relative for color prints
>
> since the paintings hues are made from pigment, not photographic dyes.
Actually, with spectral information for the pigments, the sensor, the
filtration, and the light source, it is easy to predict how the pigments will
react to light and the resulting perception of the pigments (ignoring
surround, etc).
>
Pardon my ignorance, but couldn't you just photograph Gretag's DC Color
>
Checker chart (pigments, not photographic dyes), scan the resulting chrome
>
and then simply profile this "as if" it was digital camera? I realize you'd
>
actually be profiling a *process* but assuming you use the same film, the
>
processing is stable, exposure/lighting color temp is consistent, it seems
>
you'd at least have a better starting point than the shoot/reshoot,
>
scan/rescan method.
The dyes used in the Q60 and the pigments used in the ColorChecker DC are not
the same as artists' pigments, so the result will not be the same. This is the
reason that the Q60 (and the standardized IT8.7 targets) were designed for
film scanning ONLY. The ColorChecker DC is a much better target for most 'real
world' colors, but artists' pigments are not the same. Some of them are
notorious for generating metameric problems in reproduction. The ColorChecker
DC might work for the 'well behaved' colors, but you will need to apply
special filtration techniques to tame these 'bad boy' colors (check my article
on "Color Accurate Reproduction of Artworks" on www.rmphoto.com for more
information on this).
FYI, the ColorChecker DC has a good range of color, but has been produced with
very few different pigments. One estimate I have seen is that there may only
be 11 pigments in the target. There are hundreds of artists' pigments in use.
This is why I say the use of the ColorChecker DC should not be assumed to
completely correct for reproduction problems in artwork reproduction.
Robin Myers
-----------
Robin Myers Imaging
email@hidden
www.rmphoto.com