MAKE IT RED, DAMMIT
January 18, 2004 11:37 AM   Subscribe

 
I've looked at my fair share of badly color corrected images that are generally the result of folks being color blind to some degree or just poorly toned. Most raw unprocessed images from digital cameras are extremely magenta and have that "haze" to them. If there was some sort of conspiracy I'm not exactly sure what it proves.
posted by photoslob at 11:52 AM on January 18, 2004


The imagers take monochrome pics using filters to select different wavelengths. There is no conspiracy.
posted by roboto at 11:59 AM on January 18, 2004


Ha! that's awesome stuff!

The quarter and golf ball really gave it away for me.

I remember the people at NASA repeating that it's definately a 'red' planet. I'm glad someone's hard at work debunking the whole notion of being on Mars...

but what possible motive we'd have to faking this landing eludes me. It's not like the government is averse to throwing gobs of money at projects which have no value right here on earth.
posted by Busithoth at 12:00 PM on January 18, 2004


Executive summary: NASA is altering colors in these pictures because they are trying to hide the fact that there are aliens living on mars and/or it is in fact a swell place to live.

There are a number of more mundane explanations for this, I'm sure. For example, the Pancam has an array of filters to take pictures in many different spectrums, but the individual results are Black and white -- which means the team has to combine filters and do color corrections.

I guess conspiracy is more fun.

Alien cover up conspiracies never made sense to me anyway, NASA is a government agency, they like getting money to do more stuff. Now, if they showed evidence of aliens, do you think they'd get more funding or less? Jeez, remember the stink they raised about a rock they thought might maybe have the corpse of ancient microbe?

I guess expecting logic from the tinfoil hat lot is too much to ask.
posted by malphigian at 12:03 PM on January 18, 2004


Please check to make sure your tin foil hat is securely on your head before clicking...
posted by SpecialK at 12:12 PM on January 18, 2004


changes them from an Earth-like environment into a red inhospitable environment

Yes! In actual fact, Mars is just like the San Fernando Valley! Along with altering the colors, "the people of the NASA" are using their tricksy tricks to blot out the crowds of humanoid aliens pointing and laughing at our clumsy device.

Except that... the rover never even got to Mars! Come on, everybody knows they shot those photos in Area 51. There isn't a "planet Mars," you fools, those things in the sky are just images projected from secret mountaintop bases! WAKE UP!!!
posted by languagehat at 12:13 PM on January 18, 2004


Joking and loathing aside, the Martian sky is actually believed to turn blue (even to human optics) during sunrise and I think during sunset (not sure on the latter). It'd be pretty cool to watch the sky go from pinkish-red to blue to black over an hour, and sure to prompt a surge of homesickness for future earthlings who colonize there.
posted by holycola at 12:25 PM on January 18, 2004


Mars is just like the San Fernando Valley


I knew that Lexington Steele had to be an alien or something
posted by matteo at 12:26 PM on January 18, 2004


The golf ball looks to me like the geodesic sphere at Epcot, which suggests that its new Mission: Space attraction really does take you to Mars. Wow.

Seriously, and non-conspiratorially, colorization is nothing new, although it may be questionable. For example, from a story about such photo editing, Hubble's public relations director says, "It's hard to tell the story if you don't have a stunning image to back it up. You can go out of your way to be incredibly accurate, but if people come away and haven't learned anything, then what was the point?"
posted by realityblurred at 12:33 PM on January 18, 2004


This page brilliantly reduces an entire menagerie of complex issues, including space exploration, photochemistry, government conspiracy, and impenetrable bureaucracy, right down to something I can understand: red and blue. Woohoo!
posted by scarabic at 12:56 PM on January 18, 2004


It is odd -- the calibration target is clearly there, oriented, and the blue spot is very clearly mapped to red.

What we don't know, however, is this.

1) What filters were used. The MER-A/Spirit Pancam has two cameras. Each camera has a 8 position filter wheel. One slot is empty, two are the Red Stereo pair, two are the Blue Stereo pair, and two are the Solar filters, one blue, one red. The rest range from blue to IR. Three of them in the left were chosen for true color work, at 483, 535, and 673nm.

2) What colors were mapped to what. The images from Spirit are monochrome. To make a color picture, you map a filtered image to a color channel.

This image may well have been taken in the IR.

You can get the raw frames online. However, what I haven't found yet is a listing of the image parameters -- what exposure and filter. Given those, you could combine the three-part RGB into a color picture. For the record, filter L3 is the visual red, L5 green and L6 the blue.

What I don't know about this picture is what filters were used, and what channels they were assigned to. Without that information, you can't have an opinion. I note that NASA's press release doesn't say that this image is true-color, which they usually do mention, since most of the images taken by science packages are not.
posted by eriko at 1:09 PM on January 18, 2004


Well, combining this with the new revelation that Northwest Airlines handed over passenger data to NASA it is clear that they are preparing to send all terrorists to Mars, which is quite a nice planet to live on.
posted by sebas at 1:10 PM on January 18, 2004


Or maybe NASA is making pretty red pictures to give the public something to go "ooo" at, and get more money, like they do with all the pictures from Hubble etc. Is there a problem with this?

Colour is relative anyway - how do you know my Mars red is exactly the same was your Mars red? You can't, and won't ever be able to.
posted by Orange Goblin at 1:24 PM on January 18, 2004


The site is slashdotted, err, MeFi'ed or something. Anyone have a reliable mirror?
posted by loquacious at 1:40 PM on January 18, 2004


Mars is just like the San Fernando Valley
Then they shouldn't bother trying to look for intelligent life...

I thought they'd altered the color so they could get support for more mars missions from Bush by telling him: "Look! Another Red State!"
posted by wendell at 1:44 PM on January 18, 2004


Could there be some part of a treatment from a 3D photgraph to a 2D image that would involve altering the colours? OK, so thoughts of red/green 3D glasses spring to mind, but could there be something in it? Take this picture for instance. If there are other processes, fine, but does anyone know for sure whether such a process is used sometimes, or whether the 3D photography is based on completely different technology?
posted by nthdegx at 1:57 PM on January 18, 2004


My point is, if they're not claiming the pictures are true color, and color isn't the major concern, they might be putting out 2D treatments of their 3D photos rather than images taken with a true color camera. I don't know much about photography, though.
posted by nthdegx at 2:04 PM on January 18, 2004


404 Error - File Not Found

The page or file you were looking for was not found.
posted by Steve_at_Linnwood at 2:55 PM on January 18, 2004


If you pick up almost any science text book many of the pictures from mars landers, telescopes and even the Hubble will be noted as "color corrected."

Personally I've always been disappointed that the best optics we can put on these rovers and such deliver photos that are (to my eye at least) no where near as good a the photo taken by a $3 disposable 35mm camera.

I mean - I guess the optics and cameras used in these systems are "scientific instruments" rather than simple point and shoot type cameras... so gathering data such as scale, distance and perhaps chemical make up are more important than pretty pictures.
posted by wfrgms at 3:00 PM on January 18, 2004


For those who missed it, it was mostly conspiracy blathering except for noting the curious fact that the sundial has changed color and thus the whole image must be compromised. you can compare for yourself on this rover page, where the top image shows a red patch on the sundial whereas the image at the bottom of the page (taken on earth) shows a blue patch (as well as the wire next to it). I'll admit too that that is odd but is it an artifact of the light or of how the images were processed?

What I don't know about this picture is what filters were used, and what channels they were assigned to. Without that information, you can't have an opinion.

This, and the rest of eriko's comments above, are spot on.
posted by vacapinta at 3:15 PM on January 18, 2004


Just as Jesus saves, so Google caches.
posted by mr_crash_davis at 3:42 PM on January 18, 2004


vaca:
it's not as simple as 'woo, they changed the colours to make it look more like the Red Planet of old'. Did no-one else read as far as his contrasting the highly detailed JET published pix, and the blury, dusty, reddy pix which NASA put out?
posted by dash_slot- at 4:45 PM on January 18, 2004


I thought they'd altered the color so they could get support for more mars missions from Bush by telling him: "Look! Another Red State!"

Ah, this discussion just wasn't suited to metafilter until someone decided to take completely off-topic and ridiculous swipes like that. Nice to see my faith in humanity restored.
posted by piper28 at 4:50 PM on January 18, 2004


You thinks that's bad? Check this out.
posted by electro at 4:50 PM on January 18, 2004


dash, I saw the blurry vs. clear photos but my point stands (and eriko's)

Without noting how these images were processed we cant really say which is the "right" image. Sure, the raw data can be combined to bring out detail in the rocks but if you are studying atmospheric density that may not be the best way to combine them.

RGB snaps are just snapshots of a continuous set of wavelengths. To combine them you need to make all sorts of different assumptions about the radiative spectrum of the object you are studying.

The closest approximation to "well, what would i see if i were standing on mars?" might be a true color representation but even this is just guesswork as our eyes are very sensitive to different lighting conditions, relative contrasts etc - even our own snapshots of say, sunsets, don't look like what sunsets look like to our eyes.

Anyways, astronomers don't care so much about this - they are used to seeing how things "look" in different discrete frequencies and then making assumptions (is it a blackbody?) to make sense of the data.

Meanwhile all the pictures discussed here are just "pretty" pictures that NASA or JPL has thrown out to the public. Yeah, maybe they had different rules for combining the filters but its no big deal; scientists working on this are just using the raw data.

in case anyone wants more detail, they seem to be figuring it out at this conspiracy discussion group and also note that this is nothing new.
posted by vacapinta at 6:01 PM on January 18, 2004


completely off-topic and ridiculous swipe
It's called humor, piper.
I've even made fun of liberals on an occasion or two. (But the title of that particular 'bit' earned me a top twenty ranking in Google for the term "enemas". Nobody understands my humor.)
posted by wendell at 6:03 PM on January 18, 2004


sorry, i also meant to include a link to the top of the thread, where they include a message from james bell, the Pancam Payload Element Lead for the mission:

Thanks for writing. The answer is that the color chips on the sundial have different colors in the near-infrared range of Pancam filters. For example, the blue chip is dark near 600 nm, where humans see red light, but is especially bright at 750 nm, which is used as "red" for many Pancam images. So it appears pink in RGB composites. We chose the pigments for the chips on purpose this way, so they could provide different patterns of brightnesses regardless of which filters we used. The details of the colors of the pigments are published in a paper I wrote in the December issue of the Journal of Geophysical Research (Planets), in case you want more details...


All of us tired folks on the team are really happy that so many people around the world are following the mission and sending their support and encouragement...


Thanks,


Jim Bell
Cornell U.

posted by vacapinta at 6:24 PM on January 18, 2004



What wfrgms said:

Personally I've always been disappointed that the best optics we can put on these rovers and such deliver photos that are (to my eye at least) no where near as good a the photo taken by a $3 disposable 35mm camera.


Any idea why that is, people?! Why all the fancy-Nancy filters and their associated computer interpretations? Why not just a good ol' fashioned color snap?
posted by uncanny hengeman at 8:14 PM on January 18, 2004


Why not just a good ol' fashioned color snap?

Because the trip to the chemists is a bitch?

...

I believe a monochrome camera has a higher resolution than an equivalent colour camera. Plus it's probably more scientifically usefull than something thats tuned to make things look good for our eyes.
posted by inpHilltr8r at 12:52 AM on January 19, 2004


Still no hot naked she-martian, news at 11.
posted by elpapacito at 6:17 AM on January 19, 2004


What inpHilltr8r said. Even our ole human eyeballs are much better at perceiving luminance (black and white) than we are at chrominance (colors). Today's CCDs are not much different.

This article is worth a read: Digital Secrets: How Spirit Makes Great Photos

It's good to have some debate on this: Since most other astrophotography is colored by applying filters, I trust that there's always going to be a bit of professional judgment at work, and the extent of "post-processing" probably varies from operator to operator.

Fortunately, there are other instruments aboard Spirit besides cameras... I wouldn't trust hard science to photos alone. Neither, I suspect, does NASA.
posted by skyboy at 7:15 AM on January 19, 2004


Back in the olden days, they used to use real film in the space probes. They would have a mechanical bit that would develop it, then scan it row by row to send it back gradually. Crazy.
posted by smackfu at 7:23 AM on January 19, 2004


This article, currently Slashdotted, offers an explanation by the Spirit Rover's Pancam Payload Element Lead.
posted by hyperizer at 7:35 AM on January 19, 2004


The following is from a non-metafilter member but was sent to me by a #mefi member (kmellis):

"Any idea why that is, people?! Why all the fancy-Nancy filters and their associated computer interpretations? Why not just a good ol' fashioned color snap?"—by uncanny hengeman

A film color "snap" works essentially the same way: it's made up of three emulsion layers that are sensitive to three separate ranges of frequencies of light.

Color is synthetic, it's a product of the human perceptual system. All there really is when light is reflected from an object is a reflected spectrum—so much energy at this frequency, so much at this other frequency, wherever and however closely you choose to look. A true recording of an image in the visible range would record the analog spectra of the reflected light in the visible range for each point of resolution (and would probably record the spectrum of the illuminating light as well). No current technology does this; for that matter, human color vision does not accurately measure the spectra within the range of its sensitivity. Instead, it approximates it by measuring the energy of the light in three overlapping ranges of frequencies. So, if we want to reproduce our experience of a scene in color, we only have to measure the light energy at those same three bands of frequencies. To do so may be to discard some information that might otherwise be scientifically useful. Thus, a camera like the one in question uses a variety of filters in combination such that it can record somewhat the same information that the human eye records; along with other filters and combinations to record information that the human eye does not.

Our color vision is really a beautiful natural kludge. There's no need for us to perceive a complete and accurate spectrum of an object, only its rough outlines (i.e., most of the energy is centered around a certain frequency). Our three different color receptors allow us to do two things: guess fairly accurately at the shape of the spectrum and—this is the neat part—"tag" that information in such a way that we experience a qualitative distinction between different spectra that peak in different frequencies. This makes it possible for us to process that information much more efficiently; handy when time is of the essence. It also means that our color vision makes qualitative distinctions that may or may not correspond to physical reality while, correspondingly, sometimes presenting as equivalent two very different scenes.

--kmellis
posted by Stynxno at 1:50 PM on January 19, 2004



Shite!

Thanks for all your great answers, guys. Seriously.
posted by uncanny hengeman at 9:09 PM on January 20, 2004


« Older ticker tape   |   Beyond the Fall. Newer »


This thread has been archived and is closed to new comments