Ok, all of you in this thread are right, NASA photos are good, my photos are sh*t, enjoy reddish flat skies, I'll not reply to any other post in this thread (you already decided you are right, this thread is no more useful to anybody).

Over.

__________________
-- Jumpjack --

Ok, this is the real color of Mars sky, you all wrong.
Enjoy it.



It also shows the UFO NASA was trying to hide, can you see them?... :roll:


:-?

P.S.
I obtained this just filling the sky in PhotoImpact6 using similarity 10 for "Fill" tool.

__________________
-- Jumpjack --

I understand your concern. Yes, every pixel in the sky of that picture is the same identical color. Yes, thats probably not a true representiation of the sky on Mars. I apologize if I missed the point of your original posts.

If I had to guess why they would do something like this, it is beacuse of the point I brought up earlier, which is that this is an enhanced color image. I'm betting when they enhanced the colors, the sky came out a bright blue (from the enhancement).

Rather than have people take that image, and post it online without reference to the fact that its false color, and say "See, mars really does have bright blue skies", they choose to replace the enhanced sky with the color of the real sky (yes, by obivous painting).

Or...what we are seeing is an artifact of the contrast stretching and subsequent clipping. As I said above, it is necessary to scale the images, though that scaling does create its own artifacts. The original animation you posted, with the bleeding of bright colors onto the dark terrain, exemplifies this well.

If you did the scaling as I described above, to an L7 frame, you'd end up with a blue value of no greater than the highest end of your newly scaled image. If the scaling factor were 2 or 3, that means you could only have an L7 brightness of maximum 85-128. If you created a color composite from that image, you would have twice to three times as much red in the sky than blue. But is that accurate?

The original animation you posted shows what happens when the sky maxes out the pixels. They can't go higher than 255, so they just fill the entire sky with 'max brightness'. If you left the ccd exposing twice as longthat, you'd end up with twice the brightness of the foreground and no increase in brightness in the sky, it'd just stay 255.

Now lets say you scale it back, is it fair to say that the sky's brightness no more than your 'scaled max brightness?' Just because your scaling says everything is twice as bright as it should be, it would be easy to just take the image, cut the brightness in half and go forward. But what happens when the sky should have been accumulating brightness the whole time, but was clipped to the maximum brightness before the image exposure ended? This is the nature of the clipping that has to occur. The most 'accurate' scaling doesn't tell you anything about the brightest features in the image, since they maxed out and weren't continuing to grow in value as the exposure lengthened.

Calibrating properly for sky images is amazingly difficult, especially when the camera is designed to calibrate brightness for medium bright objects, such as IMP for pathfinder and all the cameras on the MER rovers. If you have an interest in learning about the about the sky colors, I'd point you towards this image and the paper it references. It explains the pitfalls and difficulty of imaging the sky to proper contrast and color.

http://photojournal.jpl.nasa.gov/catalog/PIA01547

Hmmm... My guess for why the sky is mostly the same intensity is that the sky is quite uniform in intensity to the CCD imager in relation to the rest of the image. This means that it will cover a limited number of levels in the raw data from the CCD ADC(analog to digital converter). The sky is often the most intense feature in the images, so it will have the highest levels in the grayscale. Now, the problem is that the original image is in a 12 bit grayscale format, but the usual formats used on the internet can only use 8 bits of grayscale, this means you have to convert from 4096 different shades of gray to 256, so you will loose some information in the conversion. This is especially noticeable on the areas that have rather similar color intensity.

There are a few approaches to conversion of this type, using a linear conversion, you may map the lowest 16 levels(0-15) of the original to level 0 in the web image, then the next 16(16-31) to level 1, and so on. This creates an image that has the most correct distribution of the levels, but you will lose details. But my guess is that NASA uses a non-linear conversion that compress the high intensity areas to fewer levels, so that the detail of the midranges is improved. This would give you a uniform high intensity on the sky, but the quality of the ground features would be higher than in a linear conversion. This conversion is probably done automagicaly by some software that tries to pick the best conversion strategy for the interesting features, the rocks on the ground...

As for JPEG artifacts, you don't get them on uniform areas, as a large area of the same intensity is quite easy to compress with little or no loss...

__________________
Game over, you lose, we hope you enjoyed playing the exciting game of Thermodynamics...

Jumpjack,

People might take your concerns a bit more seriously if they didn't include thinly veiled (or outright) accusations of fraud.

There are a lot of possibilities for what you see in these images. Fakery is not necessarily the most likely of them.

So, now, your NASA hoax isn't that an image labeled "enhanced false-color" is the wrong color. It isn't that NASA removed JPEG artifacts from the sky in that image.

Your NASA hoax is that NASA exactly left unretouched a UFO in the sky, while they were busy retouching everything else.

I am in awe.

__________________
0 1 1 0 1 0 0 1 1 0 0 1 0 1 1 0 1 0 0 1 0 1 1 0 0 1 1 0 1 0 0 1 1 0 0 1 0 1 1 0 0 1 1 0 1 0 0 1 0 1 1 0 1 0 0 1 1 0 0 1 0 1 1 0 ...

It was an ironic post... :roll:

__________________
-- Jumpjack --

Maybe you could clarify your reasons for starting this thread. Is the title of the thread ("NASA hoax: why?") also meant to be "ironic"? If so, I think it would be a good idea to back off the irony a bit. It doesn't seem to be coming across as such. I think all of the egative comments you're geting are possibly due to those of us who might have mistakely thought you were seriously asserting that these touch-ups were NASA hoaxes.

Assuming an ironic use of the word "hoax": I guess I agree that NASA maybe ought to be completely open about if and how (and why) they make dramatic changes to their photos. Blocking off whole areas and painting them a particular color is a bit questionable, IMO. If there's a good reason for doing this, they ought to be clear about it.

If anything, they ought to be aware of the fact that many of those who download their images pay scrupulously close attention to them. These should not be treated as posters for grade-school classes, but as serious scientific documents. If the sky looks crappy, IOW, too bad. They ought to leave it that way or mark the photo "false color AND partial artist's interpretation".

The colors of Mars are, to a large extent, meaningless. The cameras on the rovers are not color cameras. Their purpose is to try and determine what the rocks are made of, and not what color they are. When NASA publishes a color picture, it is more or less "eye candy" for those looking at them. This matter has been discussed over and over in this forum, and has been beaten to death. The bottom line is this...NASA is not hiding anything! Give one good reason why they would.

__________________
Those who repeat History are doomed to learn it.

Because releasing a retouched photo without saying it has been retouched is a hoax, IMHO. It is retouched, not just enhanced: the photo below is retouched or enhanced?...


An image like this has the same (scientific) value of an image like this: :^o


I can't see any difference between the two images: in both cases the sky is a nonsense.

Not at all!
This exactly summarizes my opinion. Thanks for expressing it for me, I am not so comfortable with complex english speech.

__________________
-- Jumpjack --

Well, I guess I'm still more of the "artist's interpretation" way of thinking. Even so, I'm not fully convinced that this is an intentional outcome of the signal processing.

But maybe you should write to NASA and point out how much this bothers you. A tip though: don't use the words "hoax" or "faked"!

Why should they read my letter among a couple of million of others?... :roll:

__________________
-- Jumpjack --

As I suggested in my post a little earlier in the thread(Posted: Wed 12.05.2004), I think the reason(in addition to the small range of grayscale levels the bright sky is likely to have in the original image) for the uniform sky is a result of the conversion process used to make 8 bit grayscales suitable for posting on the internet, you web browser could not display the special 12 bit format it is in originally. And since you are going to loose information no matter what you do, it is logical to sacrifice the least important ranges of the grayscale, so that one get better results in the details(the rocky ground). The sky is unlikely to be painted(in the sense of someone or some program tracing the sky outline and filling it in with a solid intensity), it is more likely to be caused by the processing done by the converter software.

As for stating the reason for every thing that can be interpreted as an anomaly, lets just say that NASA could have published a detailed explanation of the entire sequence of every thing the rover can do, with full sources to their software, and people would still post about things like this.

Anyway, it is inevitable, that if NASA doesn't say something, it is seen by some as a cover-up or something and call for more information, if NASA do comment, the conspiracists say it is disinformation.

__________________
Game over, you lose, we hope you enjoyed playing the exciting game of Thermodynamics...

JumpJack, watch your tone and language. This is your only warning.

Real or not, I like the red Martian sky. Blue sky I can have anytime (if these clouds ever clear out). I'm a taxpayer and I want the red sky, Mr. NASA color faker dude.

__________________
You're a coward and a liar and a thOOF - Bart Sibrel

I think you'll find these are labelled 'Press Images' for a reason. It would be totally pointless to release serious scientific data on the internet as jpegs are you would lose all sorts of details. The Mars Rover website is simply there to keep people updated and generate publicity. The majority of people who use the site want nice pretty pictures of mars. The scientific data will be released and available for download once they've sifted through it and organised it into the PDS.

You've kinda shot yourself in the foot there. The raw images you've used (L2, L5 and L7 filters) aren't going to give you accurate colour images. You really can't use them as a basis for the martian sky being blue. Not to mention the fact that you don't have any calibration data to accurately process and balance those raw images.

There's another simple explanation for the lack of noisy pixels in the press release sky and it shows exactly why your quote "Because if you build an image from low-res or noisy images, I suppose you'll obtain a low-res or noisy image! But, in the NASA color image the noise magically disappears, and the sky is perfectly flat!" is wrong.

Take a look around the mars forums for threads related to the imaging methods used by the rovers. Many of us have explained about how it 'stretches' contrast on each colour channel to maximise the detail.

Here's an example:


In the image above I've taken a piece of sky and shown the intensity of various pixels from the image. Generally the shading looks smooth because most pixels are the same intensity. The brightest pixel on there is 192 whilst the darkest is 168

In the bottom half of the image is the same greyscale block but i've taken it into my own processing software and stretched the image. The brightest is 255 whilst the darkest is 0. This stretching increases the contrast between the pixels noticably.

The raw image you've shown is stretched. Therefore it has a lot more contrast than the final image will have. JPL accounts for colour stretching when they build their press images, and I would bet they run all kinds of algorithms and filters to clean them up. Things such as relatively small contrasts between pixels are going to be removed and the colour will be averaged out over them.

Not to mention your whole hoax theory is flawed in that JPL have posted the original pictures freely on their page. What is the point in them going to the trouble of doctoring their images only to then release the raw images?

I thought about posting this earlier so I guess I will now.

Why are you expecting scientific value from internet jpegs?


Exactly. These images are being released early to give the public something to see, as part of an effort to keep the public informed about the science results as quickly as possible, and to keep attention on the missions. However, the detailed processing of these images won't happen for some time. It just can't. There's a limit to the number of people working on the project and to the budget to pay for folks to do this stuff. And there is a requisite amount of time it takes to do image processing. And they can't even do the processing until they pull the data that tells them how long the component images were exposed and what stretch ratios were used in order to back filter the images to get the saturation levels comparable. That isn't instantaneous, either. So while it's great there are some neat pictures coming out, and it's great they're throwing raw data up so folks can do some limited work on their own, the stuff we're getting is not meant to be high quality scientific data.

Ok, sorry... :roll: I'm just annoyed by people replying to my posts without reading them.

__________________
-- Jumpjack --

Ok, now I understand the reason of "fake" pictures, but...

...but you are another person :roll: which replies to my posts without reading them (The Bad Astronomer, do you understand why I get angry?) :