Main Menu

Layers close to home

I’m trying to keep up on local affairs, and today I was reading the minutes of my local town council meeting. It’s a PDF file on the town web site. Two things annoy me about my town’s minutes: 1) They aren’t posted timely, and 2) they don’t have OCR data. As part of my looking into that 2nd issue, I opened the minutes in Adobe Acrobat and viewed properties:

image

The document was scanned by a Xerox WorkCentre 5955, and it created the PDF. As we know, some Xerox machines do amazing things, like breaking down documents into separate objects in ways that almost mimic what a human might do. This is a screen clip from the last page of the PDF file:

image

And here is the object that the Xerox created:

Minutes 9 February 2015_Page_7_Image_0003

I guess the birthers would say this must indicate a computer-generated forgery. They would be wrong.

,

22 Responses to Layers close to home

  1. avatar
    The Magic M (not logged in) July 5, 2015 at 12:53 pm #

    They would probably also complain about most parts of the signature being tack sharp but a few having heavy blur.

  2. avatar
    Reality Check July 5, 2015 at 12:54 pm #

    MRC/JBIG2 compression strikes again.

  3. avatar
    gorefan July 5, 2015 at 2:04 pm #

    There are some really cool x-ray effects in the various layers that make up the Tony Wyatt signature.

  4. avatar
    Kate July 5, 2015 at 2:10 pm #

    OMG! The forger lives near Doc. Birfers will surely be investimagatin your church. Is there a clue as to his/her real identity in the signature?

  5. avatar
    gorefan July 5, 2015 at 2:11 pm #

    BTW apparently the forger is no ordinary forger but an international forger.

    http://data.parliament.uk/DepositedPapers/Files/DEP2013-1979/Scanned_from_a_Xerox_Multifunction_Device.pdf

  6. avatar
    john July 5, 2015 at 3:00 pm #

    I agree this is troublesome to birthers to say this indicates forgery when we know that it is not. Truly impressive however, would be to see the entire name not just part of it lifted off as a seperate piece as it is shown on Obama’s BC.

  7. avatar
    Pete July 5, 2015 at 3:30 pm #

    john:
    I agree this is troublesome to birthers to say this indicates forgery when we know that it is not.Truly impressive however, would be to see the entire name not just part of it lifted off as a seperate piece as it is shown on Obama’s BC.

    john:
    I agree this is troublesome to birthers to say this indicates forgery when we know that it is not.Truly impressive however, would be to see the entire name not just part of it lifted off as a seperate piece as it is shown on Obama’s BC.

    No, john. Pieces of things are separated into separate objects in the Obama LFBC, just exactly as we see here.

  8. avatar
    Pete July 5, 2015 at 3:33 pm #

    And the knowledge of what a Xerox WorkCentre does with scans is “troublesome” to birthers’ claims in the same way that a point-blank shotgun blast to the head is “troublesome” to a rat.

  9. avatar
    Reality Check July 5, 2015 at 4:37 pm #

    If anyone has the capability of extracting and uncompressing the JPG layer I am sure that you will find the signature “YCbCr” comment that has been found in every Xerox WorkCentre scan of a color document. NBC had the tools to do this. You can also open the PDF on a Mac and resave it (in Preview).

  10. avatar
    gorefan July 5, 2015 at 6:15 pm #

    Reality Check:
    If anyone has the capability of extracting and uncompressing the JPG layer I am sure that you will find the signature “YCbCr” comment that has been found in every Xerox WorkCentre scan of a color document. NBC had the tools to do this. You can also open the PDF on a Mac and resave it (in Preview).

    Is the fact the layers are scaled 36% and 72% due to the amount of compression (still at 2:1 ratio). PDFs seem to either be 24%/48% or 36%/72%.

  11. avatar
    Dr. Conspiracy July 5, 2015 at 9:31 pm #

    The reason that the name does not lift off in one piece is because of the color variation. The slashes in the date, for example are lighter in color than the rest of the date. The part that was separated is a darker blue than the rest. The Obama birth certificate is a certified photocopy, so it’s all the same color more or less. The certification stamp was added afterwards, and is a different color.

    john: Truly impressive however, would be to see the entire name not just part of it lifted off as a seperate piece as it is shown on Obama’s BC.

  12. avatar
    W. Kevin Vicklund July 5, 2015 at 11:55 pm #

    Reality Check: If anyone has the capability of extracting and uncompressing the JPG layer I am sure that you will find the signature “YCbCr” comment that has been found in every Xerox WorkCentre scan of a color document.

    Confirmed.

  13. avatar
    W. Kevin Vicklund July 6, 2015 at 12:20 am #

    Dr. Conspiracy:
    The reason that the name does not lift off in one piece is because of the color variation. The slashes in the date, for example are lighter in color than the rest of the date. The part that was separated is a darker blue than the rest. The Obama birth certificate is a certified photocopy, so it’s all the same color more or less. The certification stamp was added afterwards, and is a different color.

    More precisely, the part in this sample that did not lift off into a separate layer was in substantial contact with the black form text. Similar to the LFBC (and many other samples). I would hazard a guess that the signature line is very thin – thin enough that it didn’t substantially alter the apparent color of the signature.

    One of the more interesting files I’ve run across contained a blue ink signature that crossed into a stamped seal (ie, inked not embossed). The Xerox MRC software changed the signature over the seal to the same black as the seal (and on the same higher-def layer), with the rest of the signature being on a blue hi-def layer. I was able to hunt down the original (I found it at work shortly after I hired on at my new job last year). On the paper original, the entire signature was blue, signed after the seal had dried.

    I’ve also found other manufacturers that do similar layer compression, such as the Canon copiers that I most commonly use at work.

    No YCbCr comment on the non-Xerox machines…

  14. avatar
    alg July 6, 2015 at 8:27 am #

    Obviously, your town council is in on the conspiracy.

    Damn! They’re everywhere!

  15. avatar
    Keith July 6, 2015 at 9:17 am #

    alg: They’re everywhere!

    He’s everywhere! He’s everywhere!

  16. avatar
    JPotter July 7, 2015 at 12:46 am #

    W. Kevin Vicklund: More precisely, the part in this sample that did not lift off into a separate layer was in substantial contact with the black form text.Similar to the LFBC (and many other samples).I would hazard a guess that the signature line is very thin – thin enough that it didn’t substantially alter the apparent color of the signature.

    One of the more interesting files I’ve run across contained a blue ink signature that crossed into a stamped seal (ie, inked not embossed).The Xerox MRC software changed the signature over the seal to the same black as the seal (and on the same higher-def layer), with the rest of the signature being on a blue hi-def layer.I was able to hunt down the original (I found it at work shortly after I hired on at my new job last year).On the paper original, the entire signature was blue, signed after the seal had dried.

    I’ve also found other manufacturers that do similar layer compression, such as the Canon copiers that I most commonly use at work.

    No YCbCr comment on the non-Xerox machines…

    In short: edge detection.

    MRC uses edge detection first, then avg color value second, to decide what is an object and what is background, then what color each object should be.

    Dark blue lines crossing black lines will either become one object or stay in the background, depending on the particular algorithm, pixel dimensions of the lines, and color contract at the intersections.

    The same picking and choosing is evident in any MRC PDF combining handwritten or typed text and preprinted line art….and the WH LFBC PDF is a perfect example of such.

  17. avatar
    john July 8, 2015 at 10:28 am #

    W. Kevin Vicklund: Confirmed.

    Yes, it would appear the Obots have stumbled across a unique signature to Xerox WorkCenters – The YCbCr comment. This may in fact be true since is hasn’t been found on any non Xerox scans. The only ones who can answer for sure would engineers at the Xerox Corporation. But, assuming for moment that this is a unique signature, there is really no way to refute this type of evidence unless we are to consider that the signature was put in there in prove that it is real document when in fact it is not to cover the deception, but I will admit that is pretty farfetched.

  18. avatar
    The Magic M (not logged in) July 8, 2015 at 1:57 pm #

    john: but I will admit that is pretty farfetched

    Colour me surprised. I thought birthers believe just about the whole world is “in on it”, including the Xerox Corporation.

  19. avatar
    RanTalbott July 9, 2015 at 11:42 am #

    john: there is really no way to refute this type of evidence unless we are to consider that the signature was put in there in prove that it is real document

    No one would have done that to “prove that it is real document”, because it wouldn’t do that.

    You can’t prove it’s a real document, because IT’S NOT A REAL DOCUMENT. IT’S A PICTURE OF A DOCUMENT.

    The Xerox evidence doesn’t prove that the PDF was made by a Xerox: it merely proves that it could have been made automatically by a Xerox, so the birther claim that it must have been created manually by a human is wrong.

    The proof that it’s a real picture is in the verification by Hawaii that its contents are correct, and the visual resemblance to the photographs and photocopies of the actual document.

  20. avatar
    The Magic M (not logged in) July 9, 2015 at 11:55 am #

    RanTalbott: it merely proves that it could have been made automatically by a Xerox, so the birther claim that it must have been created manually by a human is wrong.

    I still remember sometime in 2013 (after the A/Z presentation) how “prove the document’s anomalies can be explained by something other than forgery” morphed into “prove it’s impossible the document could have been forged”.
    So basically birthers are already way beyond that.

  21. avatar
    RanTalbott July 9, 2015 at 12:14 pm #

    For Zullups, once it was revealed that it was a two-step process, the challenge morphed into “Prove it could be done with a single press of a button”.

    I’m still hoping that one of the disappointed donors will file a mail fraud complaint, because watching them try to explain how that’s not proof that the “official law enforcement investigation” claim was a deliberate lie would be comedy gold.

  22. avatar
    bovril July 9, 2015 at 1:32 pm #

    John,

    The Obots didn’t “stumble upon” the YCbCr string they unearthed an intrinsic piece of proof as part of an involved, effective, technically ept and scientifically based investigation.

    You know, the sort of thing that Zooloops and the Kold Kustard Pussies swore blind they were doing….. yet plainly weren’t.