Slashdot Mirror
Explaining the Mars Photo Colorization
TaddyPorter writes "I've seen stories going around the 'net in regards to NASA editing photos of mars. Mainly, the sundial used for calibration showed different colors than the dial on mars. While a wide range of explanations were taking shape, the Pancam Payload Element Lead for the mission, Jim Bell of Cornell University, was kind of enough to explain the color differences."
From end of article (yes, I skipped straight there... :))
There is simply no point in adding on their site "caution these images are not 100% precisely actual colors" when no digital image is really 'actual colors'.
Quite. In fact, I'd go as far as to say that NASA expected most of the people who were scrutinising these pictures to have some experience with astronomical imaging, where almost nothing is "true" colour in that sense.
Personally, I'm in favour of as much rebalancing as it takes to make the images pretty. If they don't make full use of my eye's ability to perceive them, then what was the point of spending all that money obtaining them in the first place? So long as the raw originals are available too, who cares?
These sigs are more interesting tha
The bizarre claims of conspiracy theorists just go on and on. If you go to their websites you can read more than any brain can handle. I have read literally dozens of things that ``prove'' the moon landings were faked, for example, and each one is rather easily shown to be wrong by anyone with experience in such things.
I think the problem here is twofold: we tend to want to believe (or at least listen to) conspiracy theories, particularly to do with space. Also, the evidence is presented in such a way that, if you are unfamiliar with the odd nature of the vacuum of space and of space travel, it sounds reasonable.
>>esr>>
...doesn't NASA throw the public a bone? This color correction controversy pops up everytime a probe successfully lands on Mars and sends pictures back. One would think that they would have a standard RGB style camera for publicity shots. Chances are they can only afford to put on cameras practical for the mission, but I still believe a better solution could be provided. It probably just wasn't important to them... ;) Perhaps next time a camera could be included that features lens that provide scentific data and that can double as a publicist for NASA - spitting out RGB standard images that require no color correction.
Fun with Inkwell | www.coo
Well its OK, if Nasa wants to change the colors. No fuzz.
But can they tell if they do that and also provide pictures with alternative coloring so that the recipient have choice.
That would seem reasonable to me.
They can do that, and still Edit away all the alien artifacts...
The questions are, of course,
- if there is a tinted color light source, what would the color target display on a normal color target? What would it show via the camera with the tinted light source
- The sun is the same light source on mars as it is on earth, therefore it should be easy enough to take a solar spectrum and see what the degree of tinting is.
- With an atmosphere at 1% or less of the earth, the spectrum could nearly be the same spectrum as in a vacuum
- if the spectrums are essentially similar, then the color targets should be the same, say as on earth or in vacuum, given a clear day without dust and clouds, etc.
- Of course,there is also the matter of the end result of different photo filters getting mis interpreted. However, JPL has published some pictures with red skies, and some with blue skies, as this item from the tin foil hat crowd. This has contributed to the controversy.
See also this earlier slashdot story on the Mars SundialsSo it looks like this particular annoyance has been around for a while.
"It is a greater offense to steal men's labor, than their clothes"
Wow, do we really need another thread about article about the infamous Mars colonization. We already discussed this in the last one. It's all about filters used. When blue-according-to-the-human-eye turns extremely red, well, that's obviously when they aren't using a filter to reflect colors as seen by the human eye best, but to enhance other wavelengths. I don't really see what the problem is, and why this of all technical stuff has to be so mysterious.
The link in the article is of course slashdotted now, so here's another one explaining how a camera on the rover works:
The Panoramic Camera (Pancam)
Pay particular attention to the last paragraph there.
Beware: In C++, your friends can see your privates!
All CCDs are particularly sensitive in the IR range.
True, and the one on the Spirit Rover goes up to appoximately 1100nm.
However, that does NOT excuse the so-called "color" photos from NASA, nor does the excuse presented in the linked text. Why?
Simple reason - As the link mentions, the Spirit Rover sees the world through two identical cameras, with a set of 14 (16 with 2 pairs overlapping) narrow bandpass (around 20nm wide) color filters.
Now, it may well hold true, as per the link, that the blue paint chip looks bright at 750nm. However, WHY would they use L2 (750nm) to simulate human red vision (650nm), when they have a much closer match at L3 (670nm)? Even excusing them from doing a full spectral distribution between the seven (#8 only helps for the sun) left filters mapped onto the human visual response, they could get FAR better results just by not using L2.
And remember, despite appearances to the contrary, we do deal with "rocket scientists" here. People who should know that 670 lies closer to 650 than 750 does.
So for those who consider this a frivolous complaint by the foil-beanie-wearers, consider the utter simplicity of this so-called "problem", as well as the ease of an imperfect-but-damn-close fix (ie, use L3 rather than L2)... And suddenly it all looks a lot more strange, that a huge team of engineers failed what amounts to a 2nd-grade math test, yet successfully put a lander on a distant planet.
Furthermore, the color problem only counts as one of three major questionable points on NASA's image manipulation... If you compare the first released panorama from NASA with the version at the link below, you'll notice not only the color as wrong, but that NASA has deliberately blurred the image. Yes, deliberately - Using the wrong color channel for red doesn't account for the drastic reduction in effective resolution seen between NASA's version and the same thing generated by others from using L2, L5, and L6 (the SAME filters, still with the bad choice of L2, that NASA said they used). NASA officially claimed the haze and low visibility resulted from dust hanging in the air, yet the same image produced externally to them shows a beautiful clear sky with great visibility and no "dusty" effect.
And, I mentioned three problems, not two - The last, I consider worse than any number of failures, up to and including completely losing a mission. To cover up the "bad" colors, they CROPPED OUT THE COLOR CALIBRATION TARGET in all subsequent image releases. Yes, you read that right - Check out their web page... Early pictures they released include it (showing the wrong colors for blue and green), while the version currently on the JPL website have it strangely cut out of the picture. Sorry, but hiding the flaws in an experiment doesn't fly well in the scientific community. If they screwed up initially, they had every chance to fix it and say "oops", with no one making more than an amused comment. But instead, they chose to cover something up - Whether they did so to hide their initial error, or to hide something larger (such as water, see below), I'll leave to the beanie crowd to debate.
For those who want to see the "correct" colors, check out Keith Laney's page, which includes quite a few stunningly nice pictures.
And for those who wonder about that strange substance the Rover landed it - In full, correct color, it looks glaringly obvious that it landed in mud (yes, mud, a mixture of dirt and water).
As the article states - instead of throwing away wavelengths above the visible spectrum (as the human eye would do), they are instead clamped. Anything bright infra-red becomes bright visible-red. Net result - way too much red in the pictures.
Wow, you read the article, but apparently missed the entire point! I'm impressed.
CCDs are color blind. They take intensity maps only. Generally, they use R, G, and B filters with wavelengths as listed in the article. Many of the pictures were taken with an "R" filter that has a much longer wavelength than the usual R.
You can't "throw away" wavelength information because you don't have any. All you have are intensity maps at 3 wavelengths. You simply do not intensity maps at the middle.
If you want NASA to put out only near-true-color images, enjoy. I'll take all the other pictures and not worry so much, along with the rest of the normal humans. Of course, you'll also still have to deal with the fact that CCDs respond linearly to intensity and your eye is (somewhat) logarithmic, so any time you look at a bright source, everything will be completely wrong. Of course, everyone already knows this - pictures never look exactly the same as reality, unless they've been very very carefully taken with someone comparing the result to what they see with their own eyes, or in very controlled circumstances.
Want to know what Mars really looks like to the human eye? Go there (*). Currently, there's no other "real" practical way, without building some very expensive (and very useless) piece of equipment.
(*: You could also calculate it because you know the atmosphere and you know the input spectrum. NASA has - it's something like a yellowish-brownish-red ("butterscotch", they call it).