Fake "Star Field?" in STS-106 - NASA Manipulation Evident

reply to post by theRoboZ





Thanks Roboz,

Let me try the embed feature with your image crop:






Edit: It didn't work. Dang.


Here, I'll embed it for you:

While checking the original picture i found two other areas with cloned "stars".

Lower left quadrant, see here:

*Good eye mate.

[edit on 27-4-2010 by Exuberant1]

www.youtube.com...

this sort of thing happens a lot when using the content aware fill tool in photo-shop CS5

I just stumbled on this one, so forgive the initial somewhat uninformed response.. I'll look at it a bit deeper when I have time.. but in the meantime, a coupla comments.

They are very obviously NOT stars. Stars do NOT look like that, either on film or digital. And you can see huge amounts of chromatic aberration and subject movement in the image, and yet the tiny little dots don't show any of that. They are almost certainly 'hot' pixels of some type. Hot pixels can indeed manifest in 'arrays' and patterns, as they are read out in repeating lines, but I agree that it's an odd looking repetition..

As for cloning - I'm VERY surprised that the alleged Photoshop experts didn't wind up the gamma levels to examine the areas they are suggesting were cloned, looking for the inevitable signs in the background noise.

Cloning is very hard to do without 'smudging' the fine image noise - yet no such smudging appears. And if the 'cloner' was so incredibly incompetent as to not realise s/he was duplicating a pattern, then s/he would most certainly not be smart enough to avoid leaving the other, less easily visible, traces like that smudging effect.

But more importantly, if the areas were cloned using normal cloning techniques, the background noise should be the same around the repeated stars. Yet it is NOT.

Try it yourself - wind up the gamma level (adjust the brightness curve), and then carefully examine the repeating 'stars' - Only the 'stars' are repeated - the background patterns have subtle but quite different background variations.

Don't get me wrong, it is still just possible that it was cloned, but I would say there is less than a 0.1% chance of that.

And by the way, I will happily admit I have seen NASA busted once for very poor cloning - but the poor incompetent cloner was just trying to cover a big scratch in the film. In that case, NASA fessed up to it, and showed the original scan - there was nothing being hidden.

Anyway, I'll make some enquiries and see if I can get a bit more technical detail about the image - it seems to be an 'odd' one, not in the normal archive. BBL.


[edit on 27-4-2010 by CHRLZ]

I'm seeing a lot of the same hot pixels in other images from STS-106 from around the same time. Here are the images.

s106e5054

s106e5056

s106e5057

I've overlayed all four images in photoshop and a lot of the dots match up in each pic. Some photos don't contain all the hot pixels, but the images seem to have different exposure settings which would effect how many hot pixels show up in each image.

Originally posted by jra
I'm seeing a lot of the same hot pixels in other images from STS-106 from around the same time. Here are the images.

s106e5054

s106e5056

s106e5057

I've overlayed all four images in photoshop and a lot of the dots match up in each pic. Some photos don't contain all the hot pixels, but the images seem to have different exposure settings which would effect how many hot pixels show up in each image.

Nice work, jra (as usual). Have you found any technical details on these images? I suspect they *may* have been taken on an old DSLR, the Kodak DCS460, but I'm not 100% sure. Yet.

I have compressed the color histogram of the original image to expose artifacts and details which might be otherwise missed (similar to increasing the gamma, but more precise and selective).

What I see is multiple instances of ghosting due to refractive and reflective characteristics of the optics in the viewing instrument and/or camera used. I see no evidence of any tampering with this image as the background colours in my "squeezed" version show no signs of cloning anywhere in the image.



[edited for grammar]

[edit on 27-4-2010 by treesdancing]

I've just repeated jra's check, and yes, you can indeed see these dots repeating on the images, eg if you load:

spaceflight.nasa.gov...

and the original image:

spaceflight.nasa.gov...

..into an image editor, boost their brightness and and align them so you can flick between them, you will see they are NOT stars, they are hot pixels. It's worth noting that the DCS460 (if that was the camera used - anyone know..?) was well-known for not being very good in low light (this was in the very early days of digital cameras), and it seems this one had a lot of bad pixels. Bad pixels are often aligned along the 'readout lines' and may form repeating patterns by the nature of the way the data is read.

And it is patently obvious that in image s106e5052 they have boosted the contrast to extremes - presumably because the ISS was very dimly recorded in that pic. That contrast enhancement has revealed a LOT more of the bad pixels.

why " clone " part of the star feild at all ?

using part of the same pic - is just WTF - its just as easy to put an entire new star feild in covering the ENTIRE window / Fov

AND IMPOSSIBLE TO SPOT

as without access to EXACT details of the mission orbit - including the precice details of any pitch / yaw / roll - a layman could not determine if the star feild at any given point in an orbit is actually correct

Originally posted by verylowfrequency
I've seen many star fields that look similar to me because I'm not an astronomer. Just because they look similar doesn't mean that it has been "photoshopped". I've noticed many similar patterns in some Hubble images and yeah more than once I thought that can't be real.

They don't just look similar, they are "exactly the same"

Ok anybody think it could be stars, but the exposure on the camera is messed up and created a double ghost image of the stars. That would explain it?


Or maybe it could be dust or something, and the same effect of ghost image could happen.

Or a reflection on the window of something inside.


[edit on 27-4-2010 by _Phoenix_]

[edit on 27-4-2010 by _Phoenix_]

Originally posted by ignorant_ape
why " clone " part of the star feild at all ?

using part of the same pic - is just WTF - its just as easy to put an entire new star feild in covering the ENTIRE window / Fov

AND IMPOSSIBLE TO SPOT

as without access to EXACT details of the mission orbit - including the precice details of any pitch / yaw / roll - a layman could not determine if the star feild at any given point in an orbit is actually correct

Very true. Plus, at an exposure suitable for the ISS in sunlight, no digital (or film) camera would have a snowball's chance in hell of showing a star.

Originally posted by CHRLZ
...

But more importantly, if the areas were cloned using normal cloning techniques, the background noise should be the same around the repeated stars. Yet it is NOT.

...

I disagree,

For sure the dots are not star but hot pixel and while the "original" pixels are the same in every other photo, i think in picture s106e5052 there was some cloning.

Take the original area highlighted int the OP, as borachon noticed, it's actually repeated 3 times. Check the pixel near the bottom of the highlight box



I've boosted the levels of the image and put the 3 cloned pixel near each other, you can clearly see also the background noise is repeated.

The fact it is not pixel perfect everywhere is because, as already mentioned, they could've been cloned with different opacity so they got mixed with the original background, then the jpg compression (or re-compression) will also mangle it up.

Another interesting thing is the displacement of each cloned area is different and this makes me think it is hand made.

Edit:
Is it normal that I can't sign up for ats media?


[edit on 27-4-2010 by theRoboZ]

any chance this could be an effect of gravitational lensing? There is a case where the effect will cause a star to show up twice when the light is bent around a massive object.

www.wisegeek.com...

for a little more about the effect

If you overlay the duplicated hot pixels/stars in photoshop, you will see that they are in the exact same position, BUT they light intensity in each differs slightly and the pixelation around each individual light spot is different in hte "duplicated" area. So if the area was duplicated, it wasn't done manually but probably just be a technical artifact.

Sorry if i missed some posts but can anyone still explain why they duplicated the artifacts/stars ?

I don't understand why you would duplicated an artifact .

There's another place in the image that is duplicated as well. Can anyone find it?

I haven't read the entire thread - got busy examining the photo - so, this may have already been brought up.

Just saying. They've duped more than just those stars in this one image.

Must be the REAL space program at work. NASA is just sent up to take pictures. Then they have to edit out the REAL astronauts as they are coming and going.

reply to post by dPD89


Last one then I will go to sleep. This is a simple test in photoshop just to show why it can still be cloning.



Pure black background then add a little noise 0,2%. On a separate layer i drew some 1 pixel pencil white dots, then some 3 pixel brush white dots, then i use clone tool on them with 100% opacity (still on the same layer). then save as jpg with 50% compression. Check the result. Not 100% identical pixels and different pixelization due to compression yet they were 100% clones on the original. Imagine if the background is more complex and opacity is less than 100%.

And if you still check the original area you can see a large part with repeating background in the 3 spots.

Can anyone tell me how to embed the image here? Or it's only possible if I upload to ATS (and i can't...)

goodnight!

Edit:

Thanx exuberant1 but i tired and still got the link to external image...


[edit on 27-4-2010 by theRoboZ]

reply to post by theRoboZ


You can use imageshack.us to embed images here if you don't have an ATS media account.

Just go to imageshack, upload your image and then choose the 'forum' code provided on that page.

Yeah, its pretty much an exact duplicate. They could have used the clone stamp tool, or the patch tool circa CS5.

They need to hire better designers like me.

Wonder what they are hiding?

Here is the measured distance between the stars in the mentioned region.

[atsimg]http://files.abovetopsecret.com/files/3ad0c3416a552251.png[/atsimg]

The lower intensity of the left start on the top is simply the result of the feather of the mask used to create the patch or the softness of the clone brush.

I won't even bother about measuring the intensity values because this is clearly a retouched a image.

The visible difference could be either the options I've mentioned in my previous post of other settings on the clone brush like "flow", "opacity", etc.

But the chances of having such pattern repeating with such precision are ø.