Re: Subject: photogravures/metamerism-2
Re: Subject: photogravures/metamerism-2
- Subject: Re: Subject: photogravures/metamerism-2
- From: Robin Myers <email@hidden>
- Date: Wed, 24 Oct 2001 15:09:15 -0700
- Organization: Robin Myers Imaging
>
Metamerisms can happen with other colors too, but the effect
>
at the edges of infrared produces the strongest problems. It is
>
responsible for the greenishness of flourescent lights (and
>
extreme orangishness of incandescent) on film, the "pink violet"
>
effect when photographing flowers, and black pants turning
>
orange with some video cameras.
Again, there are many errors here. The greenishness of fluorescent
lighting is
due to excess radiation from the mercury vapor used in the lamp. Mercury
vapor
is excited and the ultraviolet light it produces excites various
phosphors on
the inside of the tube to emit visible light. Mercury vapor also has a
very
pronounced emission of green light and that is what makes them look
greenish. It
can be seen and measured even in the best "daylight" bulbs.
Incandescent light looks orange on film because the film is balanced for
a
daylight color temperature of 5500 K, which is the temperature of direct
sunlight. Incandescent lamps emit light at a temperature of
approximately 3000 K
(3400 K with tungsten halogen lamps) which is a much redder light.
Remember, low
color temperature numbers are reddish, high numbers bluish. You can also
examine
the spectra and see that an incandesent light puts out very little blue
light
compared to daylight at 5500 K, so it will appear more yellow or orange.
As for the "pink violet", this is an example of 'observer metamerism'.
This
effect is due to the different spectral sensitivities of the human
visual system
compared to the film's or the digital camera's spectral sensitivities.
The
spectral response of the object is constant and the illumination
constant, just
the human and the camera see the object differently. For all these
problematic
colors I've investigated so far, the difference has been the observer's
red
response combined with the object's red reflectance. The standard
observer's red
response peaks about 600 nm and a typical film has a red peak response
about 650
nm. The objects have high reflectance in the region from 650 nm and
higher. This
region produces a higher red response on the film than what the human
sees, thus
making a purple flower turn magenta (or pink if it is a lightly colored
flower)
and black pants turning reddish.
>
Silver absorbes all colors, including infrared, almost
>
equally, so Ansel Adams never had this problem. How is the
>
small gamut ink set doing with metamerism?
Once again, please check your sources. Black and white silver halide
film
emulsions (the commonly used ones) are actually more sensitive to blue
and
ultra-violet than the infrared. To make a good panchromatic film (good
response
to all visible light wavelengths), special techniques were developed
(look in
Hunt's book "The Reproduction of Colour" for the details). Due to the
lower
infrared response of silver halide, to make an infrared film was a
rather hard
task.
I have written a paper "Color Accurate Digital Photography of Artworks"
describing this issue of observer metamerism, its effect on
reproduction, and
it's solution, available on the Better Light website at
<
http://www.betterlight.com> under the "What's New" section and also
available
on the Robin Myers Imaging website at <
http://www.rmphoto.com>.
Robin Myers