[X] My Assistant

[SigurRosFan]

http://www.esa.int/SPECIALS/Cassini-Huygen...A7UMZCIE_0.html - A simulated view from Huygens

[deglr6328]

Question: Here is another animation of descent from a NASA release which I'm assuming we've seen before. Thing is, they claim its a TRUE COLOR image at the end of the animation (the surface image) and I definitely think it is NOT a true color image since as I understand it, DISR does not have any bayer masks or filters in front of its imaging CCDs. It only made these colorized images from fudging spectra taken by DLVS onto the image. So they can only be rough approxamations of color. Right? Someone on the wiki Huygens Probe article thinks they are REALLY true color and I can't seem to convince him otherwise......

[djellison]

Well - if you're fudging from really really good visual spectral data - then what more do you want? It's going to be more accurate than a Bayer filter anyway. I always have and always will use the word approximate before using the phrase true colour, as peoples perception of colour is different - however, I'd say it's fair to use the phrase Approximate True Colour if that's what the science team are suggesting.

Doug

[ugordan]

On a side note, it's sad how ruined the clips are in terms of video compression. The one deglr6328 mentioned and that can be found on the DISR site is 300 megabytes (!!!) and it's disastrous how blocky looking it is.
The same goes for the above ESA video, pretty cool, but the MPEG blocking artifacts are very apparent and IMHO unforgivable. You can do so much more today with an 80 megabyte file!

[JRehling]

Question: Here is another animation of descent from a NASA release which I'm assuming we've seen before. Thing is, they claim its a TRUE COLOR image at the end of the animation (the surface image) and I definitely think it is NOT a true color image since as I understand it, DISR does not have any bayer masks or filters in front of its imaging CCDs. It only made these colorized images from fudging spectra taken by DLVS onto the image. So they can only be rough approxamations of color. Right? Someone on the wiki Huygens Probe article thinks they are REALLY true color and I can't seem to convince him otherwise......

I think the Huygens spectra provide good spectral but lousy spatial resolution, so the potential is there for true-ish* color of the overall scene, and then we can paint that color over everything with only hunches as to the spatial variation. There's a very open-ended potential for trying to generate an accurately colored scene; I suppose recorded variations could be correlated statistically with different terrain types. It all suggests a monstrous effort that I guess has not yet been undertaken.

* = There are many issues with whether or not true color is possible in any situation. Those all apply here. Certainly more or less true color is possible, as Kodak disposable cameras and my new Canon digital camera seem to pull off.

[deglr6328]

Right that's my point. The spatial resolution is horrid for the DLVS so really all that can be done is to look at the spectra taken on the way down and say 'yeah that's probably ice it's seeing now......now it looks like the spectrum is of the organic sludge..maybe.....' etc. and then when you get the DISR images back you can say 'oh these chunks are likely icy bits, so let's assign this color we got from DLVS to these general icy looking areas of the image....' and that isn't really what I'd be willing to call true color. If it were a masked CCD or there were a filter wheel then yes I think we could reasonably call the images created "true color".

[Ant103]

Wow! The flash animation of the Esa is spectacular. It's seems to be trully (true?).
But, are the colors of the surface good? I don't remember that there have been color camera onboard Huygens...

[Guest_BruceMoomaw_*]

The color resolution of the combined camera/DLVS setup on Huygens wasn't THAT bad -- the DLVS mapped surface color in a kind of "checkerboard" way with about a dozen squares in a row, so when you combine its color map with the higher-resolution black and white images you get something very close to accurate for as gradually shaded a world as Titan. In fact, ASU did some test colorized images, with this combined setup, of the ASU grounds from the roof of a building, and what they got looks extremely realistic (green lawns merged in an unblurred way with gray parking lots, etc.)

[lyford]

Those images can be found here.
This is the Conic Roof Image:

[Guest_BruceMoomaw_*]

Yep, that's it. You'll notice a certain amount of color blurring along the borders, but not much even in a landscape where the color differences are really dramatic -- vastly more dramatic than Titan, where everything seems to look like a Halloween pumpkin. I think we can assume that the ESA's portrayal of colors in the DISR images is very accurate.