Additional evidence shows Russia manipulated satellite images

Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

New analysis carried out by experts working for Monterey Institute of International Studies  using satellite imagery show that Russian MoD manipulated satellite imagery with the intention to show Ukraine is behind the shot down of MH17.

The conclusion of the experts adds to many other facts showing Russia lied.  Despite many indications the photos are manipulated still a few people like award winning journalist Robert Parry believe the photos are genuine.

New York Times requested Russia for a comment. The Russian Defense Ministry said it would require three days to respond (starting Friday). The Ministry of Foreign Affairs referred questions to the Ministry of Defense.

The image below shows two satellite images: The one without the cloud was claimed by Russia to have been taken at July 14. There are a couple of objects visible.

The image showing a large cloud was taken at July 17. A couple of objects are gone.

Arms Control Wonk

The results of the analysis using specialized software were published at July 15 2016 at the Arms Control Wonk blog.

Arms Control Wonk is a leading blog on disarmament, arms control and nonproliferation. Founded in 2004 by Jeffrey Lewis, Arms Control Wonk  has since been a home to everything that is “too wonky or obscene” for publication about nuclear weapons.

Jeffrey Lewis is Adjunct Professor and Director of East Asia Non-Proliferation Program at James Martin Center for Nonproliferation Studies (CNS) at the Middlebury Institute of International Studies at Monterey.

Lewis in the past used satellite imagery for investigations like this on Suspect Defense Facility in Myanmar and North Korea missile lauchers. 

At July 15 2016, Lewis posted the results of analysis of satellite imagery presented by the Russian Ministry of Defense at July 21 2014. Russia claimed two photos prove that an Ukriane army BUK missile launcher was moved from a Donetsk military base  and two Ukraine BUK missile launchers were photographed by a satellite near a small village called Zaroshchenske.

The left photo shows military base near Donetsk ( A-1428). Right photo shows two BUKs near Zaroshchenske.

The experts analyzed the above photos using Tungstène, a suite of forensic software tools to detect alterations to images

The conclusion of the analysis is:

  • The “Cloud” Slide on the left: In terms of the image purporting to show a Ukrainian Buk surface-to-air missile launcher absent from a Ukrainian military base on the day of the shoot-down, someone appears to have modified or even added the cloud cover. Others have argued that Russia has provided a false date for the image and altered the clouds to obscure terrain features that might correctly date the image. We agree that the clouds have been altered. We cannot think of another motive for adding a cloud to this satellite image.
  • The “Two Buks” Slide on the right: In terms of the image purporting to show two Ukrainian Buk surface-to-air missile launchers, the launchers appear to have been modified or altered in the underlying satellite image. The changes could range from attempting to sharpen the launchers to digitally adding them. The presence of the two launchers is the central fact of the image; the signs of overt manipulation to this portion of the image renders it totally unreliable as evidence.

Previous indications for Russian lies

The results of the work done by Monterey Institute of International Studies adds to other facts that indicate Russia manipulated the satellite images.

  • Google Earth images show the BUK still parked at the Donetsk base at July 17 (see photo below)
  • Metabunk forum users debunked the Russian images (source)
  • Dutch satellite expert Marco Langbroek also stated the photo the photograph does not show a scene compliant with the geometric situation at 8:32 UT on 17 July 2014 as seen from Resurs-P1. (source)
  • Bellingcat did error level analysis on the photo. Bellingcat concluded the photo was faked
  • Bellingcat contacted the Geospatial Technologies Project at the American Association for the Advancement of Science (AAAS) . They concluded a significant discrepancy of azimuth deviation would indicate that either the time on the image is incorrect, the date on the image is incorrect, or both the time and date are incorrect.
    Page 32 of the Bellingcat two-year-later report has the details.
  • images taken at ground level of the Donetsk base indicate the BUK was not operational (source)
  • people living at Zaroshchenske told various reporters they did not see a BUK near the village at July 17 (more info here)
  • there is not a single eyewitness who saw a BUK missile launched or flying near  Zaroshchenske. However there are many who saw a missile flying coming from the south of Snizhne. (source)

First of all, the BUK that disappeared according Moscow from the Donetsk army base in fact was not operational.  Ukraine@war has a detailled post about this here.

This Google Earth image taken by a DigitalGlobe satellite on July 17, 2014 at 11:08 UTC shows there still is what appears to be a BUK at the Donetsk base. Lostarmour has a series of photos of that BUK here. Some of the photos were made after July 17, 2014.


Some disagree with results of Monterey Institute of International Studies analysis

Not everyone agrees with the results presented by Monterey Institute of International Studies. For example Dr. Neal Krawetz (@hackerfactor at Twitter) , which is a computer security specialist, forensic researcher, and founder of FotoForensics stated in a Tweet: “the report is a outright fraud”

I approached Krawetz by email and requested if he was willing to elaborate on his statement. The response was:

“I don’t give interviews. If I elaborate, I will blog. Right now, the conspiracy is not worth the attention.”


“Same mistakes, different software”

referring to a blog written by Krawetz in which he expresses his objection to the way Bellingcat used the  ELA system of FotoForensics software.

Another person in denial mode is US journalist Robert Parry. Parry at July 17 wrote a blog titled MH-17: Two Years of Anti-Russian Propaganda

Parry writes:

some nuclear arms control researchers at the Middlebury Institute for International Studies as definitive though there’s no reason to believe that these folks have any special expertise in applying this software whose creator says requires careful analysis

MH-17: Two Years of Anti-Russian Propaganda

The usage of the word ‘some’ suggest Parry did not do much research on the credentials of the authors of the Arms Control Wonk blogpost.

Parry also states:

In other words, the Russians would have no clear motive to doctor satellite photos since accurate ones would have shown the presence of Ukrainian Buk missile batteries in the area. You might have thought that the Times would have considered this fact relevant in evaluating claims from some amateur analysts about whether photos were manipulated or not.

Again, the usage of the word ‘amateur’ suggest Parry did not do much research on the credentials of the authors of the Arms Control Wonk blogpost.

While on the subject of being an amateur we have to conclude that in the blogpost of Parry not a single mention is done of all the evidence indicating there was a BUK parked at the Donetsk army base and there were no BUKs near Zaroshchenske.

From an investigative reporter as Parry likes to call himself we might have thought he would do a far more balanced reporting on this case.


Facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

42 Comments on Additional evidence shows Russia manipulated satellite images

  1. Charles Wood // July 18, 2016 at 11:48 am // Reply

    Without really wanting to be ad-hom, the credentials of the authors of the report are abysmal.

    They are all arts graduates mostly in political science. They haven’t the least idea about the mathematical/technical subtleties of the analysis program. They have just pressed buttons and got results that they then inexpertly extrapolate and expostulate.

    I don’t know Krawetz at all, but I’ve read his paper on detecting forgeries and I am impressed with it.

    I suggest you read it in detail before jumping to any conclusions about his abilities and reliability – in particular about his (very brief) twitter rubbishing of the armscontrolwonk report.

    • Charles Wood // July 18, 2016 at 11:58 am // Reply

      And just to make it clear. I have no opinion on the veracity of the images, but I do have a very strong opinion on the abilities of those who seek to criticise them.

      In legal-land the armscontrolwonk ‘analysis’ would not have the faintest chance of being used. It’s pure speculation by inexpert randoms equivalent to political hecklers in a crowd.

    • @Charles
      I suggest Krawetz first writes some comments on this thoughts in stead of just shouting “it is a fraud”. His statement makes it impossible to discuss.
      Just let this rest and have a look at the other evidence that shows Russia manipulated the images.

      What are your thoughts about those?

      • Charles Wood // July 18, 2016 at 12:20 pm // Reply

        Krawetz has already addressed the Bellingcat ELA ‘analysis’ in detail.

        The discussion here is mostly about the new image analysis with a bit of ‘ground truth’ appended by you. I’m not sufficiently involved to argue about the ‘ground truth’ but I can certainly criticise the image analysis.

        I’ve gone through the filters used in the Tungstene analysis and there are many obvious misinterpretations of the results. The cloning section in particular. I won’t address these here, but I may well produce a critique in future.

        • Evidence on the ground strengthen the conclusion these photos were manipulated.
          Plus the position of the satellite which likely took the image. It is safe to assume Russian MoD did not use a DigitalGlobe image.

          The image analyse results might not be THE smoking gun, combined with all the other evidence there is NO DOUBT for me Russia manipulated.
          Add this to the other manipulated evidence like the weird route change of MH17.
          Add this to the constant lies of Russia on other events like Syria.

          This all does make manipulation not just a believe it was manipulated, it is a fact.

  2. sotilaspassi // July 18, 2016 at 12:25 pm // Reply

    As we know RU lied, no reason to believe any of their productions.

  3. I have very little knowledge of image editing or manipulation, but I know enough to be able to form an opinion on the Arms Control Wonk document. Having read it, I don’t know how else to describe it except as total bollocks from beginning to end.

    There’s hardly a sentence in it that does not contain some bullshit statement. I don’t have time to go through everything they say but I’ll try to post some comment in the next day or two.

    The Russian MoD does not have great communication skills, and they’ve done a lot of damage to their own credibility in the past. In spite of that, even they should have no problem ripping the report to pieces. They might release some statement today, if what they told the New York Times is true:

    “Asked on Friday about the allegations, the Russian Defense Ministry said it would require three days to respond.”

    • It’s sad that the level of analysis related to the accident is low nowadays and it seems to be getting lower as smart people loose interest.

      • Eugene: What do you mean?

        • For you own behavior, I am disappointed that you don’t demand reasonable verification of the methods described. Similar to the verifications used in determining efficacy of medications, for example. There are so many technical words that would be used in an honest scientific analysis, which are completely absent in the dictionaries of the army of armatures driving the “analysis”.

          Brendan, I had a similar feelings while reading the key points of the DSB report. The report is written in an impeccable language, but when it comes to the really critical things, an arsenal of clever trickery is used. Not going to delve, as there is no audience to follow anyway.

          • Eugene: if you doubt the expertise of the analysts you should come forward with valid arguments.
            There are too many people claiming these experts are amateurs without providing any evidence. Use the comment function or whatever to express yourself.

            If not, refrain from commenting.

          • For one, try running their methods on other GE images. You’ll be surprised.

          • I do not have the software which was used. You do?
            So provide some examples.

          • > I do not have the software which was used. You do?

            I don’t. But don’t you think they should provide other people mans to verify their claims (and therefore provide access to the software)? Otherwise it’s a Volovik’s Elderberry bush type of proof.

            But I am pretty sure they are not going to give you the software as half of Google Earth imagery will appear to be faked according to it.

          • Eugene: it is really nonsense. I am not an export on photo analysis nor do you.
            It is up to you: you come forward with valid points or remain silent on this matter.

          • > I am not an export on photo analysis nor do you.

            You are obviously naive.

            You don’t heave to be an expert on a method to tell whether it works or not. When you go into the a chemist and buy a cough medicine do you think the only way to tell whether it works is to study all the molecular chemistry and biology and what not?

          • Eugene: the level of nonsense has reached a level which does not benefit to visitors of this website.
            Your comments are now moderated. It is up to me if your comments are made public or being deleted.

  4. From what I’ve seen, the Tungstene image analysis software does seem to be very effective at detecting alteration and manipulation. The Tungstene boss Roger Cozien also seems to be a serious scientist or engineer.

    Some people at the Monterey Institute of International Studies appear to have received good training in using the software. Jeffrey Lewis (@Arms Control Wonk) tweeted in April 2016 that they were being trained by Cozien himself. And in a reply to a joke about how to spot a fake photo, Lewis said:
    “That requires tens of thousands of dollars of software and a week-long training session.”

    Despite that training, they don’t seem to have learnt anything from it about interpreting the results. That seems clear in the case of their analysis of the Russian satellite images. They think that using the software is just a case of feeding the image into it and looking for anything that stands out in the results, such as a cloud or a Buk missile launcher.

    I’d be interested to know what Roger Cozien would say about what is revealed by those results.

  5. Daniel Been // July 18, 2016 at 1:41 pm // Reply

    But the list contains zero evidence of LIES! Just speculation, suggestion and valid question marks.

    “””Google Earth images show the BUK still parked at the Donetsk base at July 17″”””

    The same Google Earth shows more vehicles disappearing and appearing on the same place that year. So this is not evidence it was not touched on the 17th at all. Just that it’s not the BUK which was used to attack MH-17.

    “””Metabunk forum users debunked the Russian images”””

    They spotted discrepancies. That is debunking for the gullible and armchair consumers. There’s no way one can be certain about changes in soil or plant features (using that resolution and absorption spectrum) when the two sources are so different (visually pleasing often recalibrated pan-sharpened generated merge of four images vs panchromatic IR sensitive from another spy satellite). The Russian photos show way more paths and soil detail to begin with at many places, water content and so on. Way more precaution is needed here!

    “””Dutch satellite expert Marco Langbroek”””

    No he’s mainly archaeologist and amateur astrologist space buff. While I do respect his expertise he does not have any professional face to lose over it.

    “””Page 32 of the Bellingcat two-year-later report has the details.
    images taken at ground level of the Donetsk base indicate the BUK was not operational “””

    That doesn’t mean it cannot be moved around locally. Which is what the Russians claim to have spotted.

    “””there is not a single eyewitness who saw a BUK missile launched or flying near Zaroshchenske. “””
    Lack of such evidence does not debunk by itself. It’s at most indicative.

    I think the danger of group think, attention kicks, self-promotion and anti-Russian bias needs to be guarded for. We need professional people who stake their reputation to do the analysis in the end. And who allow themselves to be reviewed and criticized by peers.

    But I do have some trust in the DSB and JIT at the moment for that reason. They have probably a more neutral motive and would love to find flaws in each others analysis. That would be hopefully the most important motive since that’s their profession.

    • Daniel Been // July 18, 2016 at 1:51 pm // Reply

      Erratum: amateur astronomer, not astrologist hahaha.

      Addendum: Additional information like “images taken at ground level of the Donetsk base indicate the BUK was not operational” cannot make by itself some earlier claimed observation into a lie. It does not need to be operational to move. It’s just some indication that it couldn’t be be used for firing missiles or could drive anywhere by itself. Which was not the exact claim at that stage.

    • “””Metabunk forum users debunked the Russian images”””
      In fact Metabunk moderator Mick West did. User Herman Aven, who introduced the item, did not agree. His argument: there are technical that makes it very difficult to compare satellite images from different sources.

  6. Arms Control Wonk do not check and compare results from another photos and satellite images from Bellingcats (like Paris Match Photo, DG’s Buk satellite images or another photos from presentations). That why all that analyses is fake. It’s not analyses, it’s bellingcat’s pseudo proof.

  7. Tungstene from eXo maKina.

    Software presentation on 2015:

    Strange things from Arms Control Wonk:
    1. Do not use software on other photos to compare and check;
    2. Use only half of all available filters to check;
    3. Do not show filter’s settings (final result heavily depends on them).

  8. To some of the above posters…
    Could we please have a proper fact discussions here? Maybe this is another software used without proper scientific background, maybe it is not. Then lets discuss facts. But please, please stop all this Russia defending. It is getting so boring. As soon you discuss things that are looking bad for Russia – and yes, the total amount of discrepancies with these images doesn’t look too good for Russia – it feels like to stir up a hornet’s nest with repetitive posts by the same users again and again – counter-wise for things looking bad for Ukraine every arm-chair detective suddenly becomes a creditable source. Is this kindergarden or what? No, “Russia and the West” are not 2 football teams you’re cheering for. Real people have died. I’m sick of the overly ideological opinion majority here…

    • @Ollie: Completely agree with you although I doubt people with an agenda will take notice.
      I did block a couple of people from posting yesterday because their contributions did not add any value and the purpose of those posts seems to discredit experts without providing valid arguments.

      As someone said: fooled once: shame on you, fooled twice, shame on me.

      This website is a not a place for trolling.

  9. I cant believe the blog owner is taking this rubbish produced by Arms Control Wonk seriously.

    I’m out of here. May as well post at Bellingcrap.

    • @Billy Bob: glad you are gone seeing your contributions so far. Please stay away if you are not interested in truth but instead like to troll.

  10. sotilaspassi // July 19, 2016 at 7:36 am // Reply

    my 0.2 cents:
    As most RU sat images were easy to debunk without extra tools, the lies do not need special SW tools to prove them.
    So, the use of those “spy tools” is only “interesting” extra.

    I have not time to verify the SW code / reliability of the tools, so I personally rely on what I can verify by other means. For example by comparing to other available SAT images.

    And “fool me once, shame on you, fool me twice, shame on me” logic works for me. No need to further inspect the productions of proven liar like RU MOD.

  11. sotilaspassi // July 19, 2016 at 7:47 am // Reply

    btw. Missed this before:
    “experimental missile tracking satellite system the Russian officer was referring to. So, this specific information is incorrect.”

    Damn, so also that RU MOD info was a lie.

    (-> I’m very skeptic if US really detected BUK launch, IMO, max what they detected was rocket engine “blink” just before it ran out of fuel above clouds. Doubt high orbit SBIRS see the hot BUK engine nozzle without the flame)

    • Daniel Been // July 19, 2016 at 8:40 am // Reply

      So you decide to believe 100% the suggestion that “The description seems to refer to one of the Space Tracking and Surveillance System (STSS) demo satellites”,
      claimed by an archaeologist and amateur space enthusiast and star tracker? And around this conclusion the rest of the argument is based upon?

      You must learn a few lessons in skepticism still, I suspect. Why would it *have* to refer to STSS demo? Why not the SBIRS program? Why not checking GEO-1,2 or HEO-1,2 or 3? It’s all technically “experimental” from a certain point of view (recent, not finished, more satellites will be launched for coverage and added features).

      It’s baffling to me that people jump on the slightest question mark as proof of a bald faced lie! First they interpret through various layers, then make assumption, then start their reasoning, then “conclude” something. Typically for archaeologists to do so in their field (they suggest all the time) but in harder sciences it needs to be more rigorous and double-checked before such suspicion could even be written down.

      • sotilaspassi // July 19, 2016 at 10:12 am // Reply

        So, RU MOD + RU media deliberately do smoke screen and is not specific.

        Well, it seems it works on you.

        I try HARD to not let RU fool me again on Mh17.

  12. Arms Control Wonk make some unsubstantiated claims about what they call “the poor quality of the images” They’re not just saying that the Russian Ministry of Defense’s doctored their satellite photos, they’re also saying that the Russian MoD withheld some higher quality original images.
    “The image files are very poor quality. We are very disappointed that the Russian Federation, in such an important matter, would release such low quality images as evidence.”

    But the authors don’t offer any evidence that the images are worse than can be expected for satellite photos. They’re no worse than many images that come from Digitalglobe. In the much-publicised “Stratfor” image, dated 17 July, it’s not possible at all to make out the shape of the Buk that it was allegedly carrying.

    In the case of the Russian MoD’s two Buks in the field, the authors even claim that the MoD admitted to blurring the image:
    “according to the Russian Ministry of Defense, the field has also been blurred to hide the resolution of the satellite.”

    In fact the Russian MoD did not say that at all. They actually said that it was the Ukrainian SBU’s presentation that contained a copy with lower resolution, and that their own (the MoD’s) satellite image was high-quality:

    “The resolution of the Russian satellite image on Slide 5 has been deliberately lowered, which resulted in the outlines the terrain (i.e. field) looking smeared. The Russian Defense Ministry presented a high-quality satellite image of this area (as follows), which has no alleged inconsistencies pointed out by the SBU.”

  13. I’ll try to post more details of what’s wrong with Arms Control Wonk’s interpretation of their results later, but for now here’s a description that’s a bit less technical.

    To the human eye, of course, a big bright white sunlit cloud will appear completely different to an army base that’s beneath it and in the middle of lots of vegetation. It will also appear very different to it even after being fed through various filters in an image analysis program.

    The same goes for Buk launchers in an empty field. They would stand out from their background anyway, with or without image analysis filtering.

    Yet, the whole argument for the allegation of manipulation of those images is based on the fact that the cloud and the Buks appear very different to their backgrounds after filtering.

    If any experts disagree with that, can they say what those objects should look like in the case of an image that was not manipulated? In other words, what should authentic satellite pics of a big bright white sunlit cloud above that army base, and also a couple of Buks in an empty field, look like at the output of those Tungstene filters? If they cannot give answers to that, with details and explanations, then they are offering absolutely no evidence that the MoD images were manipulated.

    • Brendan: while I appeciate your efforts to show the expert level of the analysts, it more or less distracts from the fact where it is all about:
      was the image manipulated or not. Aside from the analysis on errors in the pic, we have two issues here;

      1. there is a BUK system missing in the image a Russian satellite took on July 17. On the same day, even within a few hours timeframe, DigitalGlobe satellite made a photo of the same spot showing a BUK. Photos taken from the ground show a BUK parked at the same spot as shown by satellite photo, being in a derelict state. Google Earth published photos taken a few days later show the same BUK in the same derelict state.

      2. There are none, as in zero, eyewitness who saw two BUKS parked at the spot near Zaro., Russia claims two BUKs were photographed.

      • If Arms Control Wonks would take their job seriously, they should have checked the satellite images provided by the SBU on the same subject. They didn’t, so they didn’t notice there was something odd with the colours.
        There are also no witnesses who saw a launch from the site west from Pervomais’kyi. “The remarkable fact is that 6-8 persons told the reporters they saw a missile which was launched south of Pervomajski”. Isn’t it remarkable that the Ukrainian experts found exactly this location in their analysis for DSB?

      • Admin, I was going to post more details about what’s wrong with the analysis, which is the subject of your article. But if you think that’s just a distraction, then OK, it’s your website.

        Regarding the two issues you mention:

        1. The Google Earth/Digitalglobe images should not be just accepted as verification. They took several months to publish the image showing the army base, by which time it was impossible to verify.

        If the GE/DG images were authentic, it’s strange that the US did not release them as soon as possible after the Russian MoD press conference. Exposing the Russians as liars would have supported American public statements about Russia covering up their interference in Ukraine.

        Also, Google finances Bellingcat. Some of its senior executives, such as Eric Schmidt and Jared Cohen, are members of neoconservative think tanks and have powerful political contacts. Google was described in a publicly released e-mail by the NSA chief General Keith Alexander as “a key member of the Defense Industrial Base”.

        As far as Digitalglobe is concerned, most of their revenue comes from US government agencies. The customer for their satellite photos of Donbass in summer of 2014 was almost certainly some American military or intelligence agency. That’s certainly true for the image of the military base A-1428 near Donetsk, since it was first published by the SBU, in a skewed and wrongly coloured form. The SBU didn’t get that picture from some purely commercial source.

        This doesn’t mean that either Google or Digitalglobe actively conspired to fake any images. Digitalglobe might not have had control of the satellite downlink since those images were very sensitive. Google and Digitalglobe might have just accepted images from some US agency and released them as authentic. But both companies are too close to the US government to ask awkward questions about information that it releases to the public.

        I’m not saying that the Google Earth images are fake, just that we should not blindly accept them as coming from an independent source. Neither would it surprise me if the Russian MoD’s photos turn out to be fake. The MoD has shown in a number of instances that it is capable of presenting false information as fact.

        It’s just that I have not seen any evidence that I consider verifiable or reliable that supports the allegations of Russian faking of images. I could go through the problems with the other “Previous indications for Russian lies” that you list above, but I don’t have an unlimited amount of time to spare.

        2. Zaroshchenske and the area around it have a small population and the field with the alleged Buks is a bit outside the village. Whether or not there really were any Buks there, nobody claims any longer that that that was the launch site. Almaz-Antey moved their calculated site several kilometers to the east or southeast.

        In the south of Snizhne, it’s very likely that some missile was indeed fired some time around when MH17 was shot down. There were many witnesses who heard or saw something, but that’s not unusual in a war zone. It could have been artillery, or else an S-A missile fired at a fighter jet flying at low altitude.

        • Bredan: I have heard better arguments in my life. This is like conspirary. Google Earth manipulated photos. Google works for US government.
          The United States did not release anything except for an impression of the trajectory of the missile which was released by the US embassy in Kiev if I am not mistaken. There must be a very good reason to not make public the evidence.

          But hey, lets forget Google and satellites. What about the broken down BUK. There are images of it and I am sure some of these were taken before July 2014.
          We also have images of the same BUK taken after July 2014, at the same position.

          How should we explain those? Let me give you some help. This could be the scenario:

          -separatists or Ukraine army stole the broken down BUK (you know who was in control of the Donetsk base in July)
          -Google earth manipulated the images or someone of SBU place another object (another BUK) and the same position where the original BUK was
          -separatists repaired the BUK
          -shot down MH17 (or Ukraine)
          -then drive back the same BUK back to Donetsk and demolished it again so it looked in the same state as July 17

          Sounds logical?

  14. sotilaspassi // July 20, 2016 at 1:18 pm // Reply

    Already in 2014 we had satellite images that proved RU MOD lied.

    And the info about the state of the BUK RU MOD said was running around shooting airliners:

  15. I looked at the report, and well they used an over saturated image of cloud, thus without noise, as a proof there was a manipulation, because SW found no noise.

    Of course that image was manipulated, as a person who is probably at 6th in the image compression, I can attest that. The manipulation provably happened when someone typed number 2 into corner of the image, and typed all that text and added all boxes. Was it manipulated another time? Well, if that manipulation happened before someone added text, it would be really funny to prove it, because damage can happen even by bad alpha blending. The small catch is, nobody is obliged to save intermediate images in JPEG, thus most methods of that analysis are simply unable to detect manipulation. (And users of that SW would definitely need a lot of more experience than 14 days of training. Doing half year of double blind analysis, or trying to invent algorithm for removal noise from image and reconstruction of original image, would make theirs analysis significantly more detailed, and they would probably reach different conclusion.)

    Anyway JPEG artifacts are horrible, and anyone who is using JPEG for military images instead of something lossless should be punished hard.×432.png I wonder when someone would try to use the red face in the middle of that image as a proof of manipulation.

    I’d rather prefer a blog post of Krawetz who has experience with image analysis, than ACW site which is biased.

Leave a comment

Your email address will not be published.