PDA

View Full Version : new here and question about MER raw images (color composit.)



Nirgal
2004-Mar-23, 02:54 PM
Hello and Hi to all ...

from a new forum member and Space/Mars enthusiast from south-west Europe (english is not my native
language, so excuse me in advance for any spelling and grammar errors :-)

OK, so here is my first question for the experts here:

As one of my Hobbies is digital image processing, I have been following
the threads about the "true" colors of Mars with great interest.
I have been experimenting with RGB-color composits from the NASA/JPL raw images (similar to the calibration process described at Keith Laney's site http://www.keithlaney.com/spirit_color_images.htm
and http://www.keithlaney.com/spirit_color_images_calibration.htm
(who BTW seems to assemble the most realistic Color composite Mars pics I've seen so far...)

However, as I understand the main difficulty with composing Color pics seems to be not only the selection/calibration of the "right" wavelength's/band filters for the R,G,B channels but rather the more important problem that each filter's individual greayscale image has been already "stretched" to the full 255 greylevel scale (in another thread I read that this is achieved by varying the exposure time)

So, the question: is there any way to "revert" the greyscale strechting
in order to obtain the "real" RGB-component greyscale raw images ?
(i.e. I susppect that, e.g. the "blue" channels L6/L7 would be much darker
in the "unstreched" original, so, after all, the more "reddish" look of the
official NASA color pics would be more accuratley reflect the "true"
surface/sky colors)

Maybe the original exposure time could be extracted somehow from the
raw image header's (are the raw EDR cubes avaiable somwhere online ? maybe at the Planetary Data System database PDS ?)

Any other ideas on how to compensate for the greyscale stretching/
varying exposure time problem in order to compose more accurate color images ?

(BTW, of course I know that it is nearly impossible to compose
"exactly true" color images, but at least I'd like to have an approximation giving us an impression as good as possible of the colors that we would really see through our own eyes, standing on the surface of Mars :)

snabald
2004-Mar-23, 05:19 PM
It seems like the pictures at http://www.keithlaney.com are pretty close. This guy seems to know his stuff when it comes to image processing. I think that Mars's landscape is much more beautiful in it's "real" colors (not the overdriven red we usually see), the ruddy browns and grays with the very light pink sky. It reminds me of Arizona's deserts... one of the most beautiful places I have ever visited.