"Anomalist/Geologist's" Test of "Real" Image Fails Under Analysis
In my quest not only for the truth about the anomalies at Cydonia, but for the general imaging excellence of our Space Program in total, no recent subject has drawn the amount of controversy as the funny little ugly image I downloaded from perhaps ASU's Themis site, maybe who knows (?) on July 25th 2002. Of course this is an extremely controversial subject, so as is likely with such, more has been put into proving something or someone a fraud than any actual observation of process using the associated datasets. In all earnesty, nobody either wants to know or knows more about this particular image than I. The reason for this is that so far I am probably the only one to have taken on the challenge of actually applying process to this image. Regardless of it's gross inconsistencies and crude appearance in comparison to what we've seen from Themis in general and the official image of the day representation, I have found it to be excellent upon full processing. Disregarding the fact that both the official and "Real" image have been thoroughly impugned by those stated to be the source as being no more than simple visual representations made for a webpage and unusable for IR color processing, this image still raises questions that must be answered by nothing less than a flood of new data from the same instrument for our further evaluation.
Out of the bad analysis resultant from very nearly all sources availably seen so far on the internet, a couple examples stand out, one of the most recent glaringly so. Written nearly everywhere there exists an anomalist or space enthusiast's hangout, Geologist/Anomalist TJ (Tripp) McCann of Mars UnEarthed (Not http://www.Marsunearthed.com, which is an excellent Mars imaging site, but the shlocky geocities copycat one) has put forth another of many by now standard examples of his regularly skewed analyses. Upon closer examination, his "proofs" in this case, much like most of his manifold "proofs" of present civilization on Mars, crumble. http://www.geocities.com/marsunearthed/mars_anoms.html
It is readily apparent that though this fellow accuses me of manipulating data, this is precisely what he has done in the case of his "study" while trying to prove I hoaxed an image..
We will now dig into the inconsistencies in his arguments.
He calls this "learning your ABC's".
I will be drawing direct quotes from his text at Communities@Anomalies.net located on this thread,
Quoted text from him will be in green, my observations will be in red text, but by suggestion I will add speaker tags for the color vision impaired
To start, he uses this image to presumably "prove" a presumed false point.
"The images on which this poll's question is founded, A, B and C, above, all have unique origins."
(KL) First thing noticed is that they are all vastly enlarged and the process used to do so is unstated.
(TM) "A - Image "A" is from the ASU official release and is from the 9.30 Ám band."
"B - Image "B" is from the Laney "REAL" TIF image, also taken from the 9.30 Ám band."
(KL) At simple first glance it is apparent that image B is NOT from the original image, which I've never sent anyone. There is a low compression high quality Jpeg copy of the original here. Compare them for yourself, and we will compare again below.
(TM) "As we all know this so-called "REAL" image is alleged to have been downloaded by Laney on July 25th by a misdirection (BAMF) process wherein he was supposedly given a more real image. This CLAIM is contradicted completely by Image "C". Image "C" is a decolorized rendition of an image posted by Keith Laney on July 30th 2002 on TEM's now defunct forum. This post was in the original thread wherein Bamf and Dan Smythe supposedly "goaded" Laney into downloading the image, which was later rephrased as accusing Dan and Bamf of having taunted Laney into processing the Image."
(KL) Immediately the bells go off. Not only is he not using the proper image, he has subjected what image he is using to some type of further processing that has added significant artifacting! Not only that, but he's not quite telling the truth about what happened in the exchanges on the BBS, which I've chronicled here,
Here is what happened when I compared his supposed band 9.30 with the actual one. He must have done some form of further processing on it, or improper resizing method, because it is most definitely not the same as the 9.30 from the original. In fact he has done something unstated that has highly pixellated the image far beyond it's original state. Bad analysis. Simply put, he started out with no other intention than to prove fraud, this is a biased way of doing what he's referring to as science.
(KL) Now we will refer to what he is calling "image C" (below) which is a product of the very first time I made an attempt at IR color processing after being both goaded and taunted (which is pretty much the same thing) as stated. I did this preliminary observation image as such and got it all wrong, shooting my mouth off a bit about it in a bit of a tiss and posting iton a forum. Well, mistakes lead to greater understanding, so I went on to learn it correctly.
In this first image I took five separate band images from the original download, fully saturated each in a different color, (red, green, blue, magenta and yellow) and then set them all at 65% transparency as color layers. These I then laid over each other and merged. Given that each of the five band image layers has different patterns of these "blocks" and each are represented as a color saturation level there is considerable overlap of these separate colors and enough blending to smooth them together.
It seems McCann has made another tremendous mistake in analysis here, because in this multiply overlain image, the tonal value of each band layer's individual color is masked or blended in both the layering process and just as much so in the conversion to grey scale. Had he done a bit more experimentation with the image he might have seen this. This effect is easily demonstrable in grey scale. Taking 5 random blocky bands from the "real" image, I layered them at 65% transparency and merged them. The results show to be nearly identical in texture to the color image blend, yet all the components were individually "blocky"
(KL) Now we see why this image is smoothed, it is not from lack of blocks on the input images, but rather a blending and smoothing caused by the layering process, rendering moot another contention point of this bad analysis. In result of his bad analysis, he also made some conclusions that are totally unrepresented given the evidence he has presented.
Then he goes on to emphatically comment based on this faulty analysis...
(TM) "What this image from Laney shows is that the impact crater southwest of the "Face" (marked by an arrow in my ABC image) is still present as an evident rounded crater with depth, and not a squared BLOTCH as that same impact crater appears on Laney's "REAL" data (image "B") "
"This image came _5 DAYS AFTER_ Laney supposedly received the special image on July 25th.. and yet given the impact crater's presence, Laney was still using what is evidently very much the same as the ASU"
(KL) What this image actually shows is that the impact and other features on the merged layered grey scale image done with the same process as image C are remarkably the same as that found on the merged layered color image. It also shows that instead of truth, Mr. McCann prefers to obscure the facts with unverifiable speculation and not much science or imaging technique
(TM)"This also proves that KEITH LANEY alone, and no other party HAD to have induced the processing artifacts in his "no-longer-real" data, as the "data" was in his custody when compromised and currupted."
(KL) It proves no such thing, but it does prove that faulty analysis and a deceptive presentation has been performed by McCann, and it could be said that he is attempting character assassination more than any type of scientific evaluation.
(TM) "Furthermore, this evidence should remove any and ALL justification for TEM/Laney and others filing under the "Data Quality Act", as it is quite clear that these parties alones are GUILTY of compromising and currupting the DATA themselves due to failed QA/QC and data controls, and grossly improper data infrared pre-processing procedures."
(KL) This is a direct misconception on Mr. McCann's part and many others, because I am not a willing part of any such "filing" under any "data quality act", nor have I "compromised" or "corrupted" any such data. Sounds like 'ole Trippy's trying to get a bit Perry Mason here doesn't it? I get a good chuckle every time I read it.
Had he been researching this subject as well as made seem he would surely have read so in the articles surrounding this image's issues found on my Mars Infrared Imaging pages here at my site and not wasted his legal talents..
An imaging consultant to NASA filing a suit about data he's getting on time and by contract! That is a pretty absurd notion, especially when I never said they had any part in it. If others wish to do so that's their business, but I fail to have ever seen any papers stating this to any effect receiving my signature. Did McCann ask me about my stance on this? Emphatically no, but still he felt free to state all the above on false premises and evident malevolent intentions.
1. The images used in his "analysis" were wrong.
2. His accusation that I destructively tampered this image is wrong.
3. The assumption that the merged 5 color image was not made from a blocky original was wrong.
4. The assumption that I am or was some part of any kind of "suit" against NASA is also wrong.
His article started from all the wrong presumptions and went south from there.
I am left puzzled.. was there anything about this man's recent analysis that was right?
well, he did do more..
(TM) The Laney induced "Blockies' that constitute the "City" when processed into the final multispectral color IR image can be seen below in these equalizations of the various IR bands from the "real" data."
(KL) Now he presents a tiny area of the 6.62 image seen here from what he claims to be the original image which I never sent to anyone.. It has to be from the jpeg at my site..
and then he enlarges it nearly 400%, equalizes, and takes it to the point of destruction, pointing out the nice contrasts in the pixellation. His image is on the left below.
This part I find amusing, because it's pretty tortuous and pushing it data wise, nevertheless I felt compelled to do the same and compare his results to mine. On the right is what I got on the very same area using the very same processes he documents using. Does it seem surprising now that his results differ? Of course not, because the proper results would have shown what I just did. Though there are some bare similarities between the two, the one made from the "real" image's band 6.62 shows far more detail, and the lineations are everything but straight line artifacts with blurred margins. He's either using an inferior image or is slanting his results, once again, bad analysis.
(TM) "What is notable about these rectinear features, blockies and lineations, is that they exclusively appear in broad, expansive areas of low detail, the planes areas. To those of you wishing to buy into the explanation that this is a "city", this may seem somehow "reasonable". However the appearance of these blockies in only exansive areas of less detail is far more simple. The Blur blurs the margines of areas of all surfaces but blurs ares of similar tone into one another. When followed by "sharpen" this makes very similar tonal areas into the *same* tone, thereby giving the impression of greater "sharpness". In
Tn Truth these rectilinear blockies and lineations ARE NOT representative real infrared signatures.. at all. NO real infrared signature of ANY FEATURE would appear as UNIFORM tonal patterns containing squared blocky areas of uniform tone and pixel-wide lines of uniform tone. THESE "artificts" seen across thesee IR bands are, in fact, characteristic of application of "blur and sharpen" and only by this means could they appear so pristine.
Beyond this, we know that the city occured AFTER the blur and sharpen admittedly employed by Laney. The image below shows the effect of burther, ADDiTIONAL blur and sharpen to the already created and existing blocky and linear pattern:
(KL) Well, let me try this, because it's already been shown that McCann is either using an inferior version of the image or is slanting results. This is the result from his own blur/sharpen experiment done on an inappropriately overdone image crop he did. Notice he shifted bands to 10.11 and chose a tiny area not delineating where on the image it is so neither confirmational or refutational analysis can be done on the actual area, 6.62 must not have done what he was trying to do with it.
(Addendum Sep 11 2003: Since the origination of this article Mr. McCann has removed his image which was posted below from the web. Tripp, you ought to put it back up, it looks worse on you). (added May 24, 2004: Of course, I did catch him promoting another just as invalid image, which is beside the blank. He doesn't have the image I use, so he's still errant, and getting monotonous in ignorance) Does any cut from what he's labeling as "Laney image" look like the copy of the original posted at the beginning of this article? NO, because it's not.
Notice directly above where Mr. McCann has removed the images he massacred (which are linked directly from his site) and replaced them with yet more scurrilous comments.
You are a simpleton liar with a bad attitude, worse yet, vulgar and a sore loser
Face it McCann, you've been outdone at your own game.
That's fine, no matter, because I have the original 10.11. I'll also choose a nondescript ROI (region of interest) and do likewise. Notice how the surface area he chose has fewer straight lines than variably angled ones, which is in conflict with his previous image observation. Regardless, the principle behind his use of this procedure should work anywhere on any band image. He's saying I introduced artifacts by repeated application of blur, sharpen and contrast. I know of no other way to prove him once again, wrong, except to use his own process and display the results below
Note: he has since added access to where it came from on the parent, although he's not reloaded the image above. Either way, the process he described is not represented by application of the filters and levels he claims to have used.
It appears that maybe Mr. McCann has applied a little more filter than he has stated here. You can see that it does affect the image slightly to subject it to repeated blur/sharpen, but the effect would have to be applied massively, which would mean more than the 3x application of each filter as shown on the last image on the right.. There is little difference between the most and least processed images, which are on far left and far right, If anything, this application has brought out more detail.
This recent "article" "BBS post" or whatever you want to call it has been one tremendous series of misstatements, bad analysis, errantly drawn conclusions, and assumptions based on shady manipulations of the data made by McCann in order to further whatever agenda he might think to assist. All the more detestable because that is the very thing he accuses me of. Repeated blur and sharpen were not applied to the original image, and as I've stated countless times before, these blocks were on the image first time I looked at it. Anyone with any imaging savvy at all knows that a 0.41!! (another error) gaussian blur is about enough to slightly soften the edges on a bright white letter, but has little effect on greyscale image clarity, even when applied multiply.
The entire lengthy diatribe he has been using to simply attempt to defame me doesn't pass even rudimentary analysis using his own stated techniques.
All this image achieved in my view is to raise a lot of questions and teach me that I have a wonderful new way to process an extraordinary new type of data. All escalations of this exploratory spirit into provably culpable accusations of conspiracy or fraud are more brought out by malicious agenda than by fact of matter, thirst for knowledge, or application of science. If it happens that NASA takes a closer look at the methods it accepts from its data vendors so be it.
The question I asked above; "Was there anything about this man's recent analysis that was right?" can be summed up in two letters..
(well, he did spell the names right)
Sorry Tripp, but your analysis on this fails worse than you at Colgate.
This article of yours is simply an act of vendetta which does not surprise me.
This in return is my style of same vendetta, to simply show you're a knucklehead.
and this new cheesy little geocities page doesn't change anything.
give to the Buy Dan a Muzzle and New Personality Foundation.
Poor Dan, he doesn't even deserve a retribution article...
Peace "Bamf", it's all good and understood.
You ought to try getting medieval where they actually DID get medieval..
Legendary Castles of the Palatinate