PDA

View Full Version : Im a little confused (Mars True Color)



Dubb
2007-Oct-18, 08:25 PM
-

Swift
2007-Oct-18, 09:12 PM
Dubb,
Sorry, I don't have an answer. But you would probably get more help if you posted this somewhere other than our conspriacy forum - maybe Space Exploration. And the mods might have a problem with so many pictures (hard on people with dial-up connections) - maybe you could post them as links.

JayUtah
2007-Oct-18, 09:39 PM
There's frequently a luminance factor you need to apply to each channel. That is the intensities in each filter wavelength aren't directly comparable.

BertL
2007-Oct-18, 09:40 PM
Sorry, I have nothing of relevance to add.

Heh, Dubb, the first time I saw the picture I thought it was a "bunch of dead fishes stranded on shore" picture.

Van Rijn
2007-Oct-18, 09:41 PM
Getting close to "true" color is tricky. Here's a good page on the subject:

http://www.atsnn.com/marscolors.html

The Backroad Astronomer
2007-Oct-18, 09:50 PM
Nasa sent along with the rover a color whell so they compare it to one the had on earth to adjust the colors.

Neverfly
2007-Oct-18, 10:14 PM
A lot of folks make this same mistake.

But they also tend to jump to conclusions and decide that Mars actually has a blue sky and those are blueberries on the ground and NASA must be lying to us and doctoring the images.

I have seen their websites. Pretty silly stuff.

The Backroad Astronomer
2007-Oct-18, 10:26 PM
Also a lot of people think that objects in space look like they do in the magazines or on tv, but those pictures have been enhanced to look better. Most objects through a telescope look grayish-greenish, at least to me.

Dubb
2007-Oct-18, 10:50 PM
-

The Backroad Astronomer
2007-Oct-18, 11:12 PM
from the article Van suggested

As Dr. Bell explained in his email, and as visible by viewing the Raw images hosted by NASA. The color-chips are not as simple as they appear. The pigments are designed to have different brightness at a variety of wavelengths. Not just RGB values. So as to "provide different patterns of brightnesses regardless of which filters we used". The blue pigment is very bright in the near-IR range. Thus the L2 plate has a very bright recording of the blue pigment.
So as Jay suggested you would have to adjust intenisity of the images from different filters, Dubb that is all I coud suggest at the moment I have never done this myself and I did not know if adjusted for the color whell.

Neverfly
2007-Oct-18, 11:30 PM
Dubb,
Are you suggesting that there is a Conspiracy?
That NASA is not being honest about the color images they release and claims about the sky color?

Or are you asking a simple question hoping to correct your error?

Van Rijn
2007-Oct-18, 11:31 PM
Here's the color wheel from Sol 173, the same day the above picture was taken. It was combined the exact same way, with the exact same filters.
Why would the color wheel look accurate,


Was it taken at the same time? Was it part of the same image? What is your reference image you used to verify that it looked accurate?



but pictures taken on the same day with the same filters have a blue sky and bluish looking stones......or Blueberries.

Images taken at different times of the day can look different, due to different lighting levels. Images taken different ways with the same filters can look different (parts of the image can be saturated). Lots of things can affect the image, even assuming you did use the correct filters combined correctly.

JayUtah
2007-Oct-19, 12:05 AM
Why would the color wheel look accurate...

I don't agree necessarily that it does. The kinds of adjustments you would make to normalize the luminosity in the terrain photos wouldn't necessarily change the color wheel photo to the point where you'd say the adjusted version is wrong. The blue cast on the left side of the gnomon knob suggests you've got too much blue in the color wheel picture as well, even thought it may not be as blatant as in the terrain photos -- although in your favor, the background behind the wheel doesn't look as blue as your other pictures.

Keep in mind that this part of the forum isn't a general question-and-answer part. So if you're innocently wondering what the exact procedure is to combine the color channels in Photoshop, you might want to request the thread be moved to the Q&A section. When you post your question here, people will appropriately assume you're proposing a hoax or conspiracy theory, the evidence of which is the inability to render subjectively natural color from specific wavelengths. That may not be what you want, and it may garner you some unwanted hostility.

NGCHunter
2007-Oct-19, 12:07 AM
Also a lot of people think that objects in space look like they do in the magazines or on tv, but those pictures have been enhanced to look better. Most objects through a telescope look grayish-greenish, at least to me.

That has to do with human vision more than true color most of the time (unless you're talking about narrow band images like hubble's - those colors are completely "false"). Strictly speaking, human "night vision" is completely color blind. Your eye uses rods for seeing in the dark and cones for color, which only get enough light to register under fairly bright conditions compared to what you see in the telescope.

The Backroad Astronomer
2007-Oct-19, 12:23 AM
NGCHunter, I do know about the night vision issue.

SLF:JAQ SFDJS
2007-Oct-19, 01:13 AM
Where's the conspiracy here? I think this thread needs to be closed immediately if the OP doesn't declare a conspiracy.

KaiYeves
2007-Oct-19, 01:22 AM
But they also tend to jump to conclusions and decide that Mars actually has a blue sky and those are blueberries on the ground and NASA must be lying to us and doctoring the images.
It's hard to do. The first Viking color images mistakenly showed Mars' sky as being blue, before the photo team realized they'd made a mistake, it was already in the papers.

Dubb
2007-Oct-19, 01:46 AM
=

Neverfly
2007-Oct-19, 01:57 AM
I declare conspiracy!

Are you declaring this in humor because SLF:JAQ SFDJS suggested to close the thread?

SLF:JAQ SFDJS has no authority here on that.

If you are saying a conspiracy is involved- Please state your case.

If not and the question was somehow misplaced- a Mod can move it.

Be warned however, if you are promoting a conspiracy theory and claim you aren't and have the Mods move the thread to another forum- That can get ugly:p

ToSeek
2007-Oct-19, 02:04 AM
Thread moved from Conspiracy Theories to Q&A

It should be noted that all of the rover images are histogram-stretched: no matter what the actual conditions, the darkest pixels are given the value of 0, and the brightest ones are given a value of 255 (or whatever the range is - I don't know if it's actually eight bits). That might be the cause of your problem here.

Dubb
2007-Oct-19, 02:31 AM
-

ToSeek
2007-Oct-19, 02:38 AM
NASA is providing images of Mars to anyone who wants them within hours of their being taken, and people still find something to complain about. Go figure.... ;)

Seriously, I'd suggest going over to the UnmannedSpaceFlight forum, probably in this area (http://www.unmannedspaceflight.com/index.php?showforum=3) and making some inquiries. There are some pretty serious amateurs over there coming up with good images. I'm not sure how they do it, but I think they'd be glad to explain, if they haven't already somewhere I just can't spot offhand.

Neverfly
2007-Oct-19, 02:44 AM
All Bold Mine:


Move it where ever you have to, or delete it of you want. The calibration methods obviously aren't known to anyone outside of NASA...

How does NASA do it?
Well, clearly high-end, purpose made image processing software is a big part of it. They also have all the relevant calibration and exposure information from the rover. quoted from http://www.atsnn.com/marscolors.html
(http://www.atsnn.com/marscolors.html)

I wish I had that magic software, because trying to fine tune the brightness or contrast on the raw images doesn't work, and doesn't remove the blue or green influence from the geology. Thats because a 601 nm filter or higher often has much less brightness in comparison to other geology around it, as you can see in photo #5. If the pebbles in that image are supposed to be red, or white or gray, then the red channel has to increase in brightness, or lower in contrast. There is no luminance channel, because we don't see that in the raw images provided. Everything to figure out what color Mars looks like from a digital camera's perspective is right there, and somehow NASA is the only one able to manipulate the pixels with their "high end software"? I don't buy it. So, you can call it a question. or a conspiracy theory, I really don't care. I'm just looking for the magic formula so that I can glimpse another world in true color. Pardon me for rubbing some of you the wrong way.

The fact that you initially posted this thread under "Conspiracy Theories"...

The fact you say "I don't buy it"

And that you say you don't care...

Very much seems to me that you believe there is a conspiracy but you don't want to be bothered to try to support the claim.

As far as rubbing the wrong way goes...
What rubs me the wrong way is someone claiming conspiracy without backing it up and vaguely trying to deny it out of a lack of desire to support the claim...

Not bothering to research the questions you are looking for because I have already found a great many answers from many hits from google including our own Bad Astronomers webpage and BAUT...

( you would have had a much better understanding of 'true color' had you read any of these articles...)

And lastly a basic uncaring attitude toward those people who would graciously offer answers to help you solve the problem.

Well, I googled it. Got lots of results. And I'm not sharing them. You can ask me nicely. :neutral:

Neverfly
2007-Oct-19, 02:45 AM
NASA is providing images of Mars to anyone who wants them within hours of their being taken, and people still find something to complain about. Go figure.... ;)

Seriously, I'd suggest going over to the UnmannedSpaceFlight forum, probably in this area (http://www.unmannedspaceflight.com/index.php?showforum=3) and making some inquiries. There are some pretty serious amateurs over there coming up with good images. I'm not sure how they do it, but I think they'd be glad to explain, if they haven't already somewhere I just can't spot offhand.


Rumor has it that Hortonheardawho is very good at this process as well.
Ive looked into him (thanks to Dfrank) and was impressed by his work too.

01101001
2007-Oct-19, 03:54 AM
Move it where ever you have to, or delete it of you want. The calibration methods obviously aren't known to anyone outside of NASA....

You kidding?

Check out unmmanedspaceflight.com (http://www.unmannedspaceflight.com/). They have a bunch of image-processing geeks there who are good with color. Some periodically get their amateur images on to the cover of Aviation Week (or similar). The MER :: Opportunity (or Spirit) sub-forum is a good place to browse.

I'm not aware that many BAUT Forum members are dedicated enough to roll their own, and therefore experienced -- except for the ones who are also at UMSF. It's not all that difficult to do as I understand it. Using older released data from Planetary Data System, I think, gives you some extra needed information, and more dynamic range too, I think, to make even better images -- but most amateurs are into doing more current images.

Member slinted here, does images at Mars Exploration Rover Imagery (http://mars.lyle.org/), but does not often visit.

People at Cornell do some nice images, too, at Pancam True Color Images (http://marswatch.astro.cornell.edu/pancam_instrument/true_color.html). They're not NASA -- though affiliated. You might enjoy the how-to (http://marswatch.astro.cornell.edu/pancam_instrument/projects_1.html) stuff there.

Michael Noonan
2007-Oct-19, 04:01 AM
I declare conspiracy!

Something fishy is going on here.
(snip)


You can still request this go back in conspiracies if you like. I wouldn't be too worried over the colouring. The door to the secret entrance probably sticks out something chronic in other hues.

Mind I would be mighty concerned about the consistent depiction of blueberries. Or you could ask it as a question (no where near as much fun) as it gives you access to more people with photo interpretation skills :)

Cheers

Toseeked :( well done 01101001 I took way to long typing.

John Mendenhall
2007-Oct-19, 07:28 PM
Here's some more:



Regardless of the color problems, thank you for the pictures. I really enjoyed them. Aren't the little impact craters beautiful?

antoniseb
2007-Oct-19, 07:38 PM
Aren't the little impact craters beautiful?
Are you referring to the little circular places where the RAT tool was used to grind into the surface?

Neverfly
2007-Oct-19, 09:19 PM
Are you referring to the little circular places where the RAT tool was used to grind into the surface?

I thought those were gopher holes... :neutral:

01101001
2007-Oct-19, 10:10 PM
Just for comparison, some other versions of the images cited in the opening post:


[Sol 173 Example] (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20173.jpg)
lyle.org Mars Exploration Rover Multispectral Color Imagery version (http://www.lyle.org/~markoff/pds/234567/1P143540537RAD3300P2401L234567C1.JPG)


1 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20173-2.jpg)
lyle.org Mars Exploration Rover Multispectral Color Imagery version (http://www.lyle.org/~markoff/pds/234567/1P143540864RAD3300P2401L234567C1.JPG)


2 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20170.jpg)
Cornell University version (http://marswatch.astro.cornell.edu/pancam_instrument/images/True/Sol170B_P2598_1_True_RAD.jpg)


3 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20179.jpg)
lyle.org Mars Exploration Rover Multispectral Color Imagery version (http://www.lyle.org/~markoff/pds/456/1P144079371RAD3336P2536L456C1.JPG)


4 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20180.jpg)
Cornell University version (http://marswatch.astro.cornell.edu/pancam_instrument/images/True/Sol180B_P2537_1_True_RAD.jpg)


5 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20184.jpg)
Cornell University version (http://marswatch.astro.cornell.edu/pancam_instrument/images/True/Sol184B_P2544_1_True_RAD.jpg)


6 (http://diffuser.us/images/Mars/Oppurtunity%20Panoramic%20Cam%20Sol%20193.jpg)
Cornell University version (http://marswatch.astro.cornell.edu/pancam_instrument/images/True/Sol193B_P2548_1_True_RAD.jpg)

Source: lyle.org Mars Exploration Rover Imagery (http://mars.lyle.org/) and Cornell University Pancam True Color (http://marswatch.astro.cornell.edu/pancam_instrument/true_color.html)

Dubb
2007-Oct-21, 08:27 PM
-

George
2007-Oct-21, 11:27 PM
Here's the color wheel from Sol 173, the same day the above picture was taken. It was combined the exact same way, with the exact same filters.

http://www.diffuser.us/images/Mars/Opportunity Sol 173 Color wheel.jpg

Why would the color wheel look accurate, but pictures taken on the same day with the same filters have a blue sky and bluish looking stones......or Blueberries.

Here is how it should look (from JPL (http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040108a.html))...

6504

01101001
2007-Oct-21, 11:48 PM
Here is how it should look (from JPL (http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20040108a.html))...

Those appear to just be early press images of the calibration target, not carefully adjusted color images of the calibration target. The soil is too blue. The blue sample chip is too blue.

Compare a more carefully done Cornell Pancam image (http://marswatch.astro.cornell.edu/pancam_instrument/images/True/Sol013B_P2548_1_True_RAD.jpg). Too bad they didn't seem to have any pure calibration images.

Ah. Here's a better version, adjusted: JPL: True Colors Shining Through (http://www.jpl.nasa.gov/missions/mer/images.cfm?id=1281)

a1call
2007-Oct-22, 12:05 AM
Why would the color wheel look accurate, but pictures taken on the same day with the same filters have a blue sky and bluish looking stones......or Blueberries.

To best of my knowledge there is no official explanation for the exaggerated blue channels.

I can offer you my conclusion however (for what it's worth :)):

The reason lies with the border-line UV filter (http://www.highmars.org/niac/education/mer/mer00.html) used for the blue channels.

Mars has an atmosphere which is less than 1% the thickness of that of Earth's.
Additionally it has no Ozone layer. It's easy to conclude that Mars' light has an abundance of UV.

Additionally a similar but milder unexpected blue effect can be seen in the images taken from the Arctic and Antarctic locales.

Just google arctic or antarctic and view the the image results. You will see what I mean.

The blue rocks and sky are for the most part high UV parts of the image showing up as blue. You would not see them if you were there watching (UV is invisible). The only exception is the blue halo around the sun which is real (Mars has reversed colors of sunset).

George
2007-Oct-22, 12:21 AM
Those appear to just be early press images of the calibration target, not carefully adjusted color images of the calibration target. The soil is too blue. The blue sample chip is too blue. Thanks. How did I miss that, but it makes me feel a little better that they missed it, too.


Ah. Here's a better version, adjusted: JPL: True Colors Shining Through (http://www.jpl.nasa.gov/missions/mer/images.cfm?id=1281) Nice. The differences between mine, Dubb's, and this are not all that signifcant, though the result is much different.

An early Viking image was adjusted to give Mars a blue sky on the assumption that the sky would actually look blue. [It has been posted in BAUT a couple of times.] They corrected it once they realized their mistake.

[Interestingly, in going to JPL's rover site, that the blue halo around the Sun in the Martian sky has been voted on as being the no. 1 image. This blue is a result of selective scattering, though the image seems to be slightly enhanced.]

Ronald Brak
2007-Oct-22, 07:41 AM
Mars is the orange planet, in English. However, other languages words for colour don't necessarily match up with the English words so in other languages Mars can be the red planet and I imagine English either borrowed the phase without checking its actual colour or simply decided that the red planet sounded better than the orange planet.

I just thought I'd mention this in case anyone was wondering.

Neverfly
2007-Oct-22, 07:48 AM
Mars is the orange planet, in English. However, other languages words for colour don't necessarily match up with the English words so in other languages Mars can be the red planet and I imagine English either borrowed the phase without checking its actual colour or simply decided that the red planet sounded better than the orange planet.

I just thought I'd mention this in case anyone was wondering.

Yeah, I'm a white guy with red hair.

neilzero
2007-Oct-22, 08:08 AM
Hi Dubb: You may want to remove the last line of your tag line as planets should decrease as the inverse cube of the distance = 30 within 50 light years. You divide by 8, if you halve the distance.
Only 2 planets with life within 20 light years as 5 cubed is 125. Neil

Dubb
2007-Oct-22, 08:16 AM
-

Ivan Viehoff
2007-Oct-22, 10:31 AM
There is an insoluble difficulty concerning what "true colour" means on Mars, or any other non-earth location. If you were to go to Mars and look at the colour wheel with your human eyes, it would look different from what it looks like on Earth, because the light is different on Mars. Likewise, a Mars rock will look different in Earth light from Mars light. (I am ignoring the numerous difficulties concerning things looking different in Earthlight according to illumination angle, intensity, surface texture, clouds, etc.)

So the procedure of trying to match the colour wheel to what it looks like on earth does not produce photos that look like what a human eye on Mars would perceive. Rather it attempts to reproduce what Mars would look like if illuminated in an earth-like light. That is probably a useful approach for science, especially for looking at rocks, since we are used to looking at rocks in earth light, and not used to looking at them in Mars light. But I suspect many do not realise that it wouldn't actually look like that to a human present on the surface when described as a "true light". In attempting to produce a "Mars light for human eye" image, probably of limited usefulness for scientific purposes but interesting for human curiosity, there are difficulties in how one would calibrate the cameras. Though there are a few "Mars light" pictures that have been published.

George
2007-Oct-22, 02:26 PM
There is an insoluble difficulty concerning what "true colour" means on Mars, or any other non-earth location. If you were to go to Mars and look at the colour wheel with your human eyes, it would look different from what it looks like on Earth, because the light is different on Mars. Likewise, a Mars rock will look different in Earth light from Mars light. (I am ignoring the numerous difficulties concerning things looking different in Earthlight according to illumination angle, intensity, surface texture, clouds, etc.) Yes, but it may not be all that much different due to our eye's ability to correct with color constancy. But, I really don't know. There is a mirror on the color wheel to allow sky illumination to be seen. This helps the color rendering process. A JPL team took over 100,000 images, IIRC, here on Earth in varying conditions to help their color rendering. It is likely they are pretty close, assuming they have had time to use all the corrections.


So the procedure of trying to match the colour wheel to what it looks like on earth does not produce photos that look like what a human eye on Mars would perceive. That is likely true but consider this. If the imaging sensors matched our eye's spectral sensitivity, then it should be easily reproducable. Of course, they don't, so corrections are needed. Just how close they can come is beyond my knowledge.

If true color was a must, there is a way to accomplish it. What we see [as color] is dependent upon the spectral energy distribution (SED) that enters our eye. Spectrometers can measure this on Mars, or anywhere. If a beam of light could be altered to duplicate this same SED, then we would see it as it woud appear to us, assuming the intensity was about the same. You have now entered the world of heliochromology! The asterocolorimeter was designed to accomplish that task, but it is only a prototype.

Kaptain K
2007-Oct-22, 11:03 PM
The "true color" is largely irrelevant. The rovers are there to do science, not to take "pretty pictures" for us to "ooh and ah" at. That it gets pictures that are "reasonably close" is just "bonus"! :clap: :dance:

George
2007-Oct-22, 11:32 PM
Yep, that point is important. Budgetary constraints alone will limit how much effort is applied for "true color" rendering; too much squeeze for too little juice is not prudent. :)

False color, of course, is even more valuable to scientists and, sometimes, artists.

Nevertheless, this might be an area that amateurs could help JPL. The more "true color" an image is, the better the representation. If the image is to represent what is really there, it is best if it can be done with accurate color rendering. WYSIWYG is preferable for the general public.