Friday, January 18, 2013

Four years later, Mike shows us how to fake "Data's Head"

        Back in 2008, in one of the very first posts to this blog, I explained how I had questioned the image that Hoagland & Bara have called "Data's Head". It's a rock in the lunar crater known as Shorty -- Station 4 of EVA-2, Apollo 17. Essentially I had asked Mike Bara, on the original Dark Mission blog, how to get from  this.... this.

        At the time (22nd May 2008) the reply I got was "you don't know anything about how images are processed."

        I guess Mike's been thinking about the problem all this time (along with writing two more books full of appalling factual errors), because today he finally responded, in the form of a step-by-step Youtube demo. Thanks, Mike. Better late than never.

        The pictorial part of the demo starts with an array of the 12 Hasselblad images shot by Gene Cernan of the inside of Shorty. The top left image has had color artificially added, and the narration soon explains why:

[01:31] One of them I've already worked on. I've done some enhancement. The head does not appear in that image, but you can see by comparison with the picture next to it that the actual real colors of the Moon are significantly brighter, significantly more contrasty, significantly more interesting than NASA would have you believe.

        Hilarious. He takes a perfectly good image, in focus and correctly exposed, showing lunar terrain which has, by nature, very little in the way of color. He then proceeds to add color, and declares that the result is more authentic than the original. This is his "more interesting" image, which he also used in Chapter 4 of Ancient Aliens on the Moon:

        The caption in the book is "Color corrected image of the lunar surface from Apollo 17 pan of Shorty crater. Note pink, purple and green hues of the entire area". His use of the term "color corrected" is totally dishonest, of course. The fakery is immediately apparent from inspecting the Lunar Rover's wheel-guards, which here appear bright scarlet. They are in fact bronze color, as you might easily see in any number of NASA shots from Apollo 16 or 17, or even from chapter 7 of AAotM. So much for "the actual, real colors of the Moon."

Image faking for dummies (like Mike Bara)

        Back to Mike's Youtube demo. He takes us through the steps, using this image--one of several that show the DH rock.

step 1: Crop to isolate the rock
step 2: Zoom in. He says that this isn't interpolation or re-sampling, but in  fact it is.

[03:10] You can already see that there is in fact ... indications of the red stripe on the upper lip, upper jaw, area of Data's Head.

No we can't, Mike. It's in your imagination.

step 3: Use Microsoft Office Picture Manager's "auto-correct"

[03:30]'s the built-in tool that makes some corrections ... it simply enhances the image a little bit -- brightens it up, does a few things to it.

It's not clear why it's legitimate to use auto-correct on a high-definition image that's already well focused and exposed. Actually, it's a way of losing data that you can never get back. Donch'a just adore the precision of that "does a few things to it"?

step 4: Color enhancement

This is where Bara steps well beyond any legitimate digital photo retouching process.  Take a look at the controls in Picture Manager:

       In the upper part of the control panel, there's an option labeled 'Enhance Color'. The instruction is "To perform the correction, click on an area in the picture that should be white". Mike does this in his demo, thus performing a second invalid corruption of the image since there is no white in the original (the brightest pixel is RGB 180,185,188.) But then, using the slider in the lower half of the panel, he ALSO slams the color saturation all the way to 100, the max (as shown in my image).

[04:22] This is nothing more than increasing, for instance, the volume of the color just the way you'd increase the volume of the audio on your TV set.

        No it is NOT,  Mike. A modern domestic audio system can reproduce considerable volume without distortion. What you're doing here is a distortion of the image and there's no conceivable justification for it.

        I invite you to load some pictures from your family album and see what effect this has. Here's what it does to the face of the master image-enhancer himself:

IS THERE ANY CONCEIVABLE WAY that image could be considered a true representation?

step 5: Using brightness/contrast controls, reduce mid-tone brightness (in the context of a professional image editor this would be known as reducing gamma)

[05:00] As you can see, it all begins to jump right out at you. [05:42] There it is, plain as day.

        Well, not really, Mike. Mike, dear, back in May 2008 you wrote "it looks like a head, it's the right size for a head, it looks like C3-PO's head. Period." Well, despite your utterly illegitimate manipulation of the NASA image, it looks nothing like C3PO.

        To me, Mike's Youtube demo is another case of what soccer players call an "own goal." In attempting to show us how legitimate the Data's Head image is, he's done the exact opposite.

        Following along with Mike's instrucs, after cropping and zooming in I arrived at a  142 x 147 image containing 1433 colors. After following Mike's color "enhancement," the same basic image contained 6084 colors. WHERE DOES HE THINK THOSE 4651 ADDITIONAL COLORS CAME FROM? Certainly not from the original. Of course not -- they were artificially added by Picture Manager. THE RED STRIPE IS IMAGINARY.

Update 2:
        I just tried it with the ultra-high-def .tiff version. Recall from my original 2008 blogpost that I special-ordered an enormous 46.1 MB, 5190 x 6175 px. tiff that was scanned direct from the archived interneg at JSC. In case any reader doesn't understand digital image formats, let's just say this image contains two orders of magnitude more information than the online jpg. Following Mike's instrucs again, here's the (significantly different) result:

        Here's an over-zoomed detail showing that the vague rusty color on the "lip" is no different from all the other rust-colored patches generated by Mike's corruptions.

Update, the wisdom of Youtube commenters again:
kalliste23 commented:
I dont' understand your "logic" - you maintain that using a different photo-manipulation proves you didn't manipulate the photo... what? Are you crazy or are you relying on your audience having a below room-temperature IQ?

Mike replied:
No, I rely on my critics having below baboon IQ's, which you obviously do. I'm surpised they even let you have access to a computer in a Turkish prison.

Ah, aspersions of homosexuality once again. What a brilliant rebuttal (NOT).


Chris Lopes said...

Let me get this straight. Bara (the self proclaimed imaging expert) is using the photo software from Office? Not Photoshop (or its open source cousin Gimp), but something from Office? That's just hilarious.

As to your actual points, yes playing with the saturation levels (or the contrast and brightness levels like Hoagie does) will get you all sorts of magical things, none of which exist in the real world. That's how film artifacts become evidence of giant glass domes and vaguely spherical rocks which lay next to a bigger rock that looks like a turkey (happy lunar thanksgiving everyone!) becomes the severed head of a beloved character from a time when George Lucas made good movies. Bara is either too stupid or too dishonest to see this.

astroguy said...

Two things I would add. First, I have *never* addressed Data's head. If anything, there's a possibility I mentioned it in passing in one of my podcasts about Hoagland and pareidolia. So Mike's stuff about me in there is gratuitous.

Second, towards the end, Mike says that all that color noise in there are JPEG artifacts. Except for the red stripe. That's real. Sure ...

Chris - as to the software, I'm going to give Mike the benefit of the doubt here in that he was trying to show that you can do this with the most rudimentary image tools that the majority of people would have. We know he has Photoshop based on all his screengrabs from the ziggurat stuff.

Trekker said...

What amuses me is that these Apollo photos from the despised NASA are genuine, and totally to be relied upon, whereas the excellent quality LROC photos from the despised NASA are fakes, and not to be relied upon. What gives?

jourget said...

Even if you assume that the original photo was manipulated, how in the world can he claim that he's perfectly reversed the original manipulations, instead of just messing the photo up further?

Tara Jordan said...

I guess I am too sarcastic and cynical for the debunking game. The way I see it,Bara is trying to make a living,cannot really blame him for trying.People like Bara only exist because there are legions of gullible idiots willing to be deceived.Social Darwinism.....

Trekker said...

He's still trying to debunk pareidolia, Expat, in the comments to the second version of his video. Did you see that?

Chris said...

Do you realise we've still got all this shit to come around again for Mars? A new book, more alien space junk strangely familiar to Bara against all odds, more 'enhanced' oversaturated Martian images and magenta/cyan jpeg ringing artifacts being presented as real. The joys.

Mulvaney said...

I left a comment on the YouTube site, but am uncertain if Bara will approve it to be posted. But as I understand it, he is stating that by manipulating images you can make them even more real. It is a little like stating that that the world becomes clearer on psychoactive drugs. But I can enhance anything to make it unrecognizable.

Bara states that any 6th grader could do what he did. Perhaps. I know a few sixth graders who enjoy manipulating images. But I would not call it research or trust the results

Anonymous said...

I left a message on his YT, and he hasn't posted my response to his reply to me..In fact I think he's deleted several comments from what I saw yesterday. I think it's because (in my case) I used logic, common sense and my own experiences of image manipulation to make my point. I wonder if Mike approaches all things that contradict him in this way? He drops a cup and breaks it, will he insist it isn't broken, or that it's only perception that says it's broken? I'm curioous as to what he means by 'organically enhanced' that images that don't use pesticides?

expat said...

I just did an important update, using the specially-ordered .tiff image, with far more information than the jpgs Bara worked with. Take a look.

Jiminy Oddbird said...

Photographs are not scientific evidence. Science requires an experiment that can be duplicated with the same results by all competent researchers. Garbage in/garbage out. NASA controls the data. Therefore, it cannot be independently verified. By the way, what is the temperature on the Moon, at what temperature do mechanical cameras cease up and fail to function, and how cold can photographic film get, and still render an actuate representation of color? Even in black and white, it's still a curious photo.

expat said...

Here's a page on that.

I assume by "cease up" you really mean "seize up" -- right?

Anonymous said...

As someone who uses Photoshop in my profession, and who has been doing so for over a decade, I can say without hesitation that Mike Bara knows absolutely nothing. The way he just cranks up the saturation on a image makes me shake my head. I'm surprised he doesn't put his fraudulent pictures through a ton of filters to further "enhance" them. What a joke this Bara guy is.

Anonymous said...

that "red stripe" that bara sees is the exact same color as the surrounding terrain. So if it as bright read as he thinks it is, he might as well color in all the terrain bright red as well. Its a damned rock.

expat said...

Anon: Thanks for your contribution. Obviously, I couldn't agree more. I really think I proved the point with my trusty .tiff (for which I paid good money, I may say).

expat said...

Anon: What do you make of the rather obvious rectangular patches in the Hoagland image? They're at 45° because he rotated the image after his overlay process.

FlightSuit said...

Bara seems to be relying on the fact that a lot of people watch TV shows like CSI and believe what they see in that show, that computers can magically tease out of a photograph additional details that aren't readily apparent in the original.

As to why Bara is using Office's image editor, instead of Photoshop's, that's no mystery: When you have a lease payment on a stripper-filled BMW to worry about, it's really hard to justify buying a copy of Photoshop.

Trekker said...

Have you seen Mike's new tattoo? Any idea what the Hebrew writing says?

Anonymous said...

I have a question. He's white balancing using (what appears to be an overexposed, which is a no-no, but I digress) a white patch on the top of the rock.

Then, in the final image (as produced by Hoagland), that same patch has a greenish cast.

If he really believes it to be white (as suggested by his choosing it to re-balance what was more than likely ready balanced by the scanner operator), wouldn't he look at the final image and say "whoops, I just hosed this picture."

I'm betting he could recreate the same results by balancing with the area that becomes red.

expat said...

There wasn't a white patch until he used auto-correct, which "does a few things to it".

The RGB of that spot in the ultra-high def tiff file is 230,230,220. In the online jpg, the brightest pixel is 180,185,188. There is no white pixel -- or even any "almost white" pixel -- anywhere in the picture. Thus, Mike Bara's use of the 'Enhance color' feaure is thoroughly invalid.

James Concannon said...

I'm no great genius at image editing but it's obvious to me that this demo shows incompetence.

As you say, expat, a decisive "own goal."

Unknown said...

Love it, Expat.

Jiminy Oddbird said...

The camera's might have been capable of working if they had been contained in a heated enclosuer, but the picture of the astronaut wearing the camera attached to his chest, is unprotected from the cold temperatures of the Moon. Unless, of course, the Moon isn't as cold as NASA has claimed.

Anonymous said...

Misti, you are out of your depth. Please show us your thermal analysis of the camera internal temperature. Show your math with error statements for every parameter.

Gaby de Wilde said...

I see a rock shaped like a head. I'm not confused, this is all I see.

I can make all sorts of fun hypothesis about it, in no particular order:

1 - god put it there to fool scientists.

2 - it is the head of a robot

3 - nasa thought it would be funny to put a head shaped rock on the moon set

4 - it's just a rock!

What hoagland did to the images both increases the quality of the image but it also reduces it, it depends how you look at it.

Saturisation isn't an unfair way to get at the colors. Indeed if you do it to multiple images you get a nice red lipstick, heck I even managed to extract a bit of chin. What you also get is artifacts from each layer. I examined some of the more obvious box borders (bits) and by rotating those they also contribute to a more head shaped shape. Trying to correct the artifacts I eventually ended up with a properly shaped nose with 2 holes in it. The left one might as wel be prouced by overlapping shadows from different angles, it looks awesomely detailed as if someone drew a nose on the image, which technicall is precicely what I did. I can for example "fix" the artifiacts on the forhead into a bunch of lines just as easily as into alien writing.

Shifting the layers allows me to make it look like a head but I'm chosing to do so. I was also able to stack them in such a way that the right side of the face just doesn't match the left side.

Right has a nice circular eye socket, the right has a lowered brow. Right has a low cheek bone the right side has a cavity there.

Hoagland is trying to make things as much like artifical objects as possible, it is what he does. His evidence is not in one of these uhhh creations but in the amount gathered.

It seems dubious but this is why we bother making pictures on other worlds. I admit the odds are terrible for this rock to be a head. But find 2 or 3 of these heads, each with the same feature, then the story changes entirely.

Your debunking then becomes valuable research :D

Either way,

Thanks for the articles it was funny.