Xerox/WH PDF – Henry Blake Affidavit – Part 17 Resolution

Our friend Hermitian made a somewhat puzzling suggestion when I pointed out to him that the Whitehouse PDF had been significantly compressed.

First of all he failed to understand that the 150 DPI resolution for the background was only part of the loss in quality of the background, after all, the JPEG compression itself had reduced the size of the image significantly but at the expense of about 50% of the quality.

Then he suggested that I should print out the document at 600 or 1200 DPI…

Does he not understand that such a picture will not truly be 600 or 1200 DPI? Anything above 150 DPI will be interpolated and this will result in a softening and blurring of the image. Anyone familiar with information theory would know that you cannot recreate lost information by just blowing up the resolution.

It’s quite fun to help Hermitian understand these basic principles of image analysis.

17 thoughts on “Xerox/WH PDF – Henry Blake Affidavit – Part 17 Resolution

  1. Had Obama been born IN the White House,it would not help him. A President is required,in addition to being born in the USA , to have been born of Parents who were BOTH Americans.

  2. Had Obama been born IN the White House,it would not help him. A President is required,in addition to being born in the USA , to have been born of Parents who were BOTH Americans.

    You are wrong. Natural Born means born on US soil.

    There we go again… such unfamiliarity with our legal history and the Supreme Court rulings.

  3. “NBC

    “gsgs placed his findings here and provided us with many useful tools to analyze his findings.”

    Where is GSGS ? He’s been silent for months.

    [NBC: He has better things to do I am sure]

    Where are these useful tools that he provided ?

    Provided where?

    [NBC: See below…]

    NBC and GSGS posted that GSGS wrote native code to extract his results directly from the WH LFCOLB PDF file.

    So where’s are the codes?

    Funny that no one seems to have a link to the codes that he claims to have written.

    [NBC: You never bothered to search my friend… That’s the problem… A diligent researcher does not give up too easily. I will give you a hint Here]

    So he extracted his (x,y) and (W,H) values directly from the PDF file without using any other information?

    [NBC: All the information is in the PDF once you extract the raw object]

    Or without using the high level graphics programs such as Adobe Illustrator, Adobe Acrobat, Adobe Photoshop, or InkScape?

    So he only used a MAC OS with Preview and the PDF file?

    For Hermetian’s benefit I will copy gsgs’s observations because he appears to be confused about gsgs’s findings and his observations that they do not align with 16 bit boundaries.

    So you are finally admitting again that GSGS’s stuff doesn’t satisfy the 16 MOD 0 condition?

    [NBC: That’s what he and I have always stated. Weird..]

    After all GSGS had stated that as one of his original findings.

    [NBC: I am very familiar with that… So why are you now suggesting otherwise?…]

    And you were so anxious to climb aboard his bandwagon back then that you parroted his failure for the 16 x 16 blocks.

    [NBC: It was not a failure, it was a data point that needed to be explained and of course it was. You have a strange understanding what makes a failure. A failure is saying that x cannot create something and then you are show otherwise🙂 It happened to you several times recently. ]

  4. “NBC

    “He was right, the PDF file contains exactly those instructions to rotate, expand and invert. Why a forger would work with rotated objects is something the ‘researchers’ have never explained… This is typical for most claims of forgery… No plausible or credible explanation is given.”

    Gee I thought you said that Preview took care of all those weird rotations and inversions in the PDFs created by the Xerox forger.

    [NBC: You have not been paying attention. But I showed you what happened when Preview touched the document. And no, the rotations remain, of course… But Illustrator is no longer confused.]

    “GSGS

    “But, of course the human forger could have rotated and inverted it after the work was done so to confuse the birthers “

    So the Birthers didn’t have to explain it because GSGS already did.

    [NBC: That’s a pretty ad hoc explanation my friend. No wonder your claims are going nowhere… You must not understand the sarcasm used here]

    Bingo GSGS ! That’s exactly what the forger did.

    [NBC: How do you know? You still have no coherent explanation, just ad hoc suggestions that show that the forger is both incompetent and a good forger…Surely even you must understand that your explanation cannot even compete with the one provided by the Xerox workflow…]

  5. I provided my answers above inline as Hermitian’s inability to properly use blockquote led to me doing it myself. Seemed like a good moment to insert my comments.

    Needless to say, Hermetian’s responses failed to impress.

  6. “NBC repost of GSGS

    “the upper right corner becomes upper left in the original rotation. All 9 layers are rotated by 90° counterclockwise and color inverted in the WH-pdf. I assume that they were always handled in this format by the software. Then there must be some additional instructions in the pdf how to rotate,invert,expand and overlay the 9 layers. This is presumably encoded in one of the 3 additional bitstreams, I haven’t studied the pdf-format yet.”

    “And spin your partner all about…”

    I thought that GSGS said that all those rotations and color inversions were in the PDF…

    [NBC: Yes, he is correct]

    So he didn’t study the PDF? But I thought that he wrote the code from scratch that extracts all the data for his analysis from the PDF?

    [NBC: All you need is the right tool to extract the commands that paint the imagemasks on top of the jpeg]

    And some of the data has to be extracted from bitstreams ?

    [NBC: Which is trivial if you have the right software]

    So some of the data is in binary content streams that are embedded within the PDF?

    [NBC: You are late to the party my friend. What do you think a jpg is or jbig2 or a monochrome bitmap… You have no clue, do you as to how a PDF works…?]

    Boy this is getting deep. If you know what I mean.

    [NBC: Yes… It makes a lot of sense right now. Your unfamiliarity with PDF’s raw format runs quite deep.]

    But guess what? Some of the data that GSGS needed for his analysis of the object boundaries IS in the embedded binary streams.

    So again — where are all these magic tools that GSGS created?

    [NBC: He uses tools like pdfimages and shows a dump of the raw PDF so I am sure he is using many of the same tools I am using but he is working on a Windows machine. So he uses irfanview instead of GIMP]

  7. I am amazed that it took this long for Hermitian to realize that the PDF contains countless binary streams… Wow… where do I start…

  8. I think the point that needs to be made over and over and over is that PDF would be the last file format a real forger would choose as an original file. A real forger would work in a native format for an photo editing program and they convert it to a PDF or JPG in one of several ways. The most secure way would be to print a copy and scan to PDF, which of course would leave no electronic traces at all.

    When poor Adrien Nash at my blog finally came to the realization that the AP JPG could not have come from the LFBC PDF he had to invent a mysterious high resolution PDF as the source for both documents. These guys are hung up on PDF’s.

    Sterngard Friegen at the Fogbow has a great saying that Orly Taitz is practicing cargo cult law. What Adrien and Henry are practicing is cargo cult image analysis.

  9. Where do I start… A PDF has a well defined layout although there are an almost infinite number of ways a particular text can be rendered. This makes PDF versatile and offers many additional hints when looking at the raw data since software leaves tell tale signs.

    The way a PDF rendered works is to immediately read the end and find the offset to the xref table.

    A PDF file contains the following:

    the header
    the document body, containing at least four indirect objects
    the xref cross-reference table
    the trailer (including startxref and %%EOF).

    The header contains the PDF version information and a high bit string to determine the endianess of the file.

    The four indirect objects are /Pages /Catalog /Page and a stream object and make up a single page minimal PDF.

    Much more here which presents an introduction to PDF format.

    Note that PDF files contain streams, which are often filtered resulting in many binary streams that are not easily accessed without applying the correct filters.

    FlateDecode is the simplest compression tool and uses gzip like encoding but lacks the header information so standard gzip/gunzip tools may not work.

    Other filters include

    /ASCIIHexDecode /LZWDecode /ASCII85Decode /RunLengthDecode /FlateDecode /DCTDecode and many more.

    In most cases a PDF contains binary streams and the highly compressed WH PDF is no exception.

  10. I have another question which I may turn into a real blog posting.

    True PDF layers only became a specification from version 1.5 onward. So my question is: If you take an illustrator PSD which contains various layers and save to PDF, what is saved to the PDF?

    While I can understand that Illustrator may read ImageMasks as ‘layers’, I am interested to see what Adobe does when such elements are created and printed to PDF.

    Hermitian, finally a situation where your Illustrator tool has some relevance in the investigation…

  11. I was somewhat shocked that Hermitian expressed surprise that the PDF contains binary streams… Yes, this is not something one typically encounters in Illustrator but somehow I had assumed some familiarity with the PDF format.

  12. I am amazed that it took this long for Hermitian to realize that the PDF contains countless binary streams… Wow… where do I start…

    1 or 0

  13. NBC

    NBC · July 23, 2013 – 02:55 ·

    “I am amazed that it took this long for Hermitian to realize that the PDF contains countless binary streams… Wow… where do I start…”

    How about loading the PDF file “birth-certificate-long-form.pdf” into your favorite text editor and show us exactly which lines of text contain the (x,y) offsets for the lower-left corner of each of the eight non-backgroud layers from the origin location in the lower-left corner?

    And then what does my question have to do with the bitmap streams embedded within this file?

    [NBC: Perhaps our friend is not aware that the much of the data are stored in PDF in streams which may or may not be encoded in a compression format. FlateDecode is the most common one for basic compression. There are tools beyond a text editor to show these data. In fact, for a researcher a hex editor would be better. Of course, the best is to use a data decoder which renders the information legible to all of us. What about loading the PDF into your favorite editor and you report on what you see? It’s trivial to do, is it not?]. Direct objects contain <> markers and are directly readible and are often followed by a stream which encodes a particular aspect of the object.

    The easiest object to located is the background as it is encoded as a jpeg (DCTDecode). Let me cut and paste the relevant part.

    7 0 obj
    <</Length 8 0 R /Type /XObject /Subtype /Image /Width 1652 /Height 1276 /ColorSpace
    11 0 R /BitsPerComponent 8 /Filter /DCTDecode >>
    stream
    

    I am sure that a capable researcher can extract the width and height from these data?

    My data dump shows

    obj 7 0
     Type: /XObject
     Referencing: 8 0 R, 11 0 R
     Contains stream
    
      <<
        /Length 8 0 R
        /Type /XObject
        /Subtype /Image
        /Width 1652
        /Height 1276
        /ColorSpace 11 0 R
        /BitsPerComponent 8
        /Filter /DCTDecode 
      >>
    

    A bit more legible… Enjoy looking at the raw data, there is so much hidden in it..

  14. How about loading the PDF file “birth-certificate-long-form.pdf” into your favorite text editor and show us exactly which lines of text contain the (x,y) offsets for the lower-left corner of each of the eight non-backgroud layers from the origin location in the lower-left corner?

    As I showed, the height and width of the objects are described in the Obj itself. However the offset, scaling and rotation come from the actual rendering command which contains the various Do operators.

    The object in question is Obj 4 0

    4 0 obj
    << /Length 5 0 R /Filter /FlateDecode >>
    stream
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
    endstream
    endobj

    Which translates into after applying the ZLIB filter to decompress the FlateDecode encoded stream.

    ‘q Q q 18 14.40002 576 763.2 re W n q 0 -792.96 612.48 0 -0.24 792.48 cm /Im1\nDo Q /Cs1 cs 0.1059 0.17650 0.1216 sc q 0 -348.96 436.56 0 89.28 581.28 cm\n/Im2 Do Q 0.34510 0.3922 0.3529 sc q 0 -47.76 186.72 0 304.56 108.96 cm /Im3\nDo Q 0.302 0.34510 0.3216 sc q 0 -10.08 65.76 0 170.16 89.76 cm /Im4 Do Q\n0.2549 0.3373 0.2627 sc q 0 -29.52 54.72 0 440.4 274.08 cm /Im5 Do Q 0.3412\n0.4353 0.3412 sc q 0 -11.28 51.84 0 103.44 254.88 cm /Im6 Do Q 0.2549 0.3373\n0.2627 sc q 0 -8.16 16.8 0 349.68 322.08 cm /Im7 Do Q 0.9412 0.9725 0.9216\nsc q 0 -58.32 52.08 0 176.16 185.76 cm /Im8 Do Q /Cs2 cs 0.9647 sc q 0 -31.68 34.08 0 251.76 784.8\ncm /Im9 Do Q Q’

    The stream should look familiar because I have provided it several times. If you need an explanation as to what it does, let me know. Oh yes \n is just a meaningless ‘line feed and can be replaced with a proper character such as a space or newline’

    I reformatted it to make it more readable

    'q Q
     q 18 14.40002 576 763.2 re W n 
     q 0 -792.96 612.48 0 -0.24 792.48 cm /Im1 Do Q
     /Cs1 cs 0.1059 0.17650 0.1216 sc 
     q 0 -348.96 436.56 0 89.28 581.28 cm /Im2 Do Q
     q 0.34510 0.3922 0.3529 sc
     q 0 -47.76 186.72 0 304.56 108.96 cm /Im3 Do Q
     q 0.302 0.34510 0.3216 sc
     q 0 -10.08 65.76 0 170.16 89.76 cm /Im4 Do Q
     q 0.2549 0.3373 0.2627 sc
     q 0 -29.52 54.72 0 440.4 274.08 cm /Im5 Do Q
     0.3412 0.4353 0.3412 sc
     q 0 -11.28 51.84 0 103.44 254.88 cm /Im6 Do Q
     0.2549 0.3373 0.2627 sc
     q 0 -8.16 16.8 0 349.68 322.08 cm /Im7 Do Q
     0.9412 0.9725 0.9216 sc
     q 0 -58.32 52.08 0 176.16 185.76 cm /Im8 Do Q
     /Cs2 cs 0.9647 sc
     q 0 -31.68 34.08 0 251.76 784.8 cm /Im9 Do Q
     Q
    

    The q … Q combination encapsulates the cm operation to the Object in question. /Im1 is the jpeg /Im2 to /Im9 are the 8 monochrome bitmaps.

Comments are closed.