Janus van Helfteren wrote: > Hi List , Spectral reflection is something that we suffered from a lot > with film , (certain coloured fabrics coming up different colours on > film/print than they looked to the naked eye) .
and then Thomas Holm wrote: I'd prefer to call it observer metamerism (the same object appearing different to two observers, human and camera/film). > > Has anyone got any ideas on how to 'cure' it , probably in terms of in > camera cures , as batch correction play havok with skin tones . What you describe sounds like a problem I came across a few years ago when some green objects recorded as a rich sepia on film. Eventually we pinned it down as "Anomalous infra-red Reflectance" and was due simply to the fact that the human retina responds to a certain range of wavelengths whilst films record a slightly different range. Specifically, bog-standard colour films are sensitive further into the red end of the spectrum than our eyes. Mostly this isn't a big deal, but some artificial green dyes also reflect a large proportion of far- (but not quite infra-) red which we don't see but the film does, resulting in green + red = something you weren't expecting. The problem then is that you can have two green fabrics that look identical but you can't predict how they'll record photographically. (The job I was shooting at the time was for a company who were launching a super-improved new range of zips for dress-making and the two important points that had to be made were 1) they were much better than the previous products and 2) they were the identical colour as the ones they were replacing. Mostly they were. Except for the green.) There's another problem at the other end of the spectrum where ultra-violet will make some dyes fluoresce; they absorb the invisible u-v and re-emit it as a visible colour. This often gives wedding photographers lots of fun. How film and digital cameras differ in these circumstances I don't know and it'll depend on the spectral sensitivity range of the CCD. As far as a "cure" goes I don't think you'll be able to do much on camera. If you try to filter out the far-red you'll most likely stuff up the colour overall and building a profile for a non-constant problem sounds tricky. I guess you may have to spend yet more time in front of the computer. I suppose you could shoot on 10 x 8 transparency and then have it re-touched... Not really helped, I suppose, but anyway, Regards Martin Farquharson =============================================================== GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE
