| So,
How Do We Know It's the Right Color?
by Frank Snively (AVAS Co-President) - 1/19/04
CLICK
PHOTOS FOR MARS ROVER WEBSITE
The Surface of Mars
Pancam Calibration Target
Note from the editor: The object above is mounted on the
Spirit & Opportunity Mars Rovers. It combines a sun dial,
a commemorative plaque AND various calibration targets,
which include color targets used to adjust the color settings
of Spirit's and Opportunity's panoramic color camera. For an
article from JPL on this, see: http://marsrovers.jpl.nasa.gov/
mission/spacecraft_instru_calibr.html. - JMB |
The very recent successful
landing of “Spirit” on Mars, and the pictures that are
coming back are very exciting to see, but they do beg a fundamental
question. How are we to know that the orange dust and brown rocks
and pink sky are indeed correct? Would we see the same thing if
we were standing there?
Here on Earth it is
relatively straight forward to verify things, most of the time.
We take a picture (or a video if we are more modern). We process
it. We look at the print or at the screen, and either we look at
the original scene or try to recall what we saw. If it doesn't match,
either the computer or the film processor has to go back to work.
But, what about a
picture on Mars. Certainly no one has been there. Or, a bit closer
to home, how about the many dazzling color pictures of nebulae and
gas clouds and other astronomical objects? If we look at the object
in question through a telescope - not that anyone is going to let
us look through one of Caltech’s 8 meter telescopes in Hawaii
- all we see is gray with faint coloration. The objects are just
not bright enough to permit the human eye discern colors.
Let’s answer
the question for Mars first. The first lander, Viking, went there
about 25 years ago. The people at JPL knew well that the atmosphere
was very thin, and dust storms, which surely blew very high had
been observed many times. Certainly there was no guarantee that
the gray, yellow and brown observed in Earth’s dust storms
would also be seen on Mars. Also the planetary geologists wanted
as accurate a rendition of the color of Martian rocks as possible.
The method chosen
by JPL was to send along a color chart. And a very carefully selected
color chart. They knew, from such things as Roman and Greek mosaic
tiles and various kinds of old pottery found around the world, that
some kinds of porcelain retained color indefinitely despite severe
environmental conditions. So they selected various color tiles,
and subjected them to particle bombardment, radiation exposure,
ultraviolet, x-rays and what have you, to verify that the colors
were indeed stable in conditions of exposure to space environments.
When Viking landed and the cameras began to work, the raw images
were corrected so that the color charts were correct. And that meant
that the rest of the colors were correct, also. Yes, the pink sky
is “for real”.
The scientists at
JPL had prior experience from Caltech to rely on, and some methods
to adapt to their own use. In the 1950’s there was no such
thing as a color film which would “track” the correct
colors if the object was so dim that a 1 hour exposure was needed.
(For the record, the problem is called “reciprocity law failure”.)
Even in the prime focus cage on the 200 inch telescope at Palomar,
something like the ring nebula in Lyra still looks mostly gray to
the eye. In a joint effort with Ansco, the problems were worked
out, and some beautiful color photos were taken. What was done was
to make up color charts and special dim light illuminators and attenuators.
Ansco’s test film was exposed under the same conditions as
would occur on the telescope (the 200 inch is f/3.5 at prime focus,
so test pictures were taken at f/3.5 for the expected exposure time).
Then the chemists at Ansco adjusted emulsions, dyes, developers
and all the other “black art” tricks of the film maker
until the test charts came out right. They did one other thing as
well. There is also colored sky glow, same idea as aurora only much
dimmer, as well as scattered light pollution. To take out the background
effect, each exposure of an interesting object was accompanied by
a picture of nearby sky, dark except for scattered stars. The sky
background was subtracted from the object picture.
Not only with Mars,
but with other pictures, the calibration and correction process
continues today. The imagers in the Hubble Space Telescope have
calibrators on board to make sure the focal plane arrays are responding
as intended in their particular wavelength band. In the visible,
the calibrators match the color vision characteristics of the eye,
so the pictures are “correct”.
Today, with color pictures
from Hubble Space Telescope or from Mars on the TV evening news,
it is easy to forget that there was a time when color pictures were
rare and wonderful things. Today, as then, there has been a determined
effort to define what “accuracy” means and to make sure
that the pictures are indeed accurate. |
