Posted on 08/07/2013 6:29:11 PM PDT by Seizethecarp
The following image is a composite created by scanning the WH LFBC using Xerox WorkCentre 7655 upside down using the automatic feeder. The resulting file was opened in Preview, the image rotated 180 degrees and printed to PDF. The resulting PDF was opened in preview, the layers unlocked and moved to the side. In addition, a close up of the signature was ‘blown up’ to show how the background layer, not surprisingly, has filled in some of the white that resulted from the separation of the background and foreground layers.
Note how for example the signature block is fully separated.
(Excerpt) Read more at nativeborncitizen.wordpress.com ...
Thanks. I also get the part about the haloes. It’s the talk of algorithms, Mixed Raster Compression and the like that I’m unable to follow well. In the end, I think that the report by Hayes will be damning enough all by itself, in its handwriting/font analys.
Is Guthrie’s physically bigger than Applewhite’s photo?
“Guthrie’s ‘document’ is physically bigger.”
Could that be because the LFBC PDF has a white margin around the border? Even after you release the clipping mask. The actual security paper should be like the Guthrie photo - no white border with the security pattern running from edge to edge.
Below are NBC’s responses to butterdezillion:
http://nativeborncitizen.wordpress.com/2013/08/15/butterdezillion-questions/
Butterdezillion questions
Posted on August 15, 2013 by NBC
Butterdezilion: I’m scrambling like crazy to get a million things done. So if you or anybody else can double-check my understanding of this, it would be very helpful.
NBC: I will see what I can do
Butterdezilion: Number substitutions happen on what machines? Do they happen on the default setting for the machine, or do you have to lower the quality and file size to get number substitutions? Did Xerox set up warnings for the user to know that substitutions could occur with those particular settings?
NBC: Yes, there were warnings but not always obvious. Xerox released a list of affected scanners
Butterdezilion: What settings are required to create an automatic clipping mask that takes off the outer edges of the scanned document? What settings create multiple layers, and are the layers created reproducible each time, or are they seemingly random? For instance, you lay the paper on the bed of the scanner. Then you press scan 5 different times, one right after the other, changing nothing and not even opening the lid of the scanner. Do all 5 scans give you the same layers – a reproducible result you would expect from a machine carrying out exact algorithms in a controlled setting?
NBC: No, they do not repeat between scans even if you leave the document on the scanner. The layers are not random as the obvious ones repeat: Signature, date stamp(s). The algorithm is susceptible to minor variations in how a color is interpreted and once a difference happens, its effects can cascade across the object. The settings are the ‘default’ setting on the machine but they can be changed by the administrator I presume. In some instances, the Xerox actually created the same mysterious speckled images that so far had remained unexplained. It is actually a known issue. Good thing is that its effects remain mostly invisible to the human eye.
Butterdezilion: If you save a scan on the 7655 without previewing it, is the saved file different than if you preview it on the 7655 before saving it? If you send the scan directly in an email is it in PDF format? If they open it on a PC will they be able to move around pieces of content? If they open it on a Mac will they?
NBC: Yes, they are very different. The Xerox created file has a lot of differences with the final document, most notably, the JBIG2 compression. Since Preview saves it to a lower PDF standard (1.3) it cannot create the same file. The resulting files however look extremely similar at the PDF object level. The ability to move around objects depends on the software. Illustrator does it both on Mac and Windows.
Butterdezilion: If you import a digital file to the copier does the copier make layers out of a digital file as well, before printing it?
NBC: I am not sure if the copier supports this workflow. When printing, the computer creates a printing stream and typically does not send the document across the network. To a printer, there are just no layers. Remember that the foreground objects are just bitmasks that get printed on top of the background.
Guthries is even bigger than the image after the clipping mask is removed. And the two ‘dots’ on the right side of the backgroup with the clipping mask do not show up in her ‘photo’.
All three are physically different and can easily be verified by the cross hatch pattern.
But clipping mask was applied to create an 8 1/2 by 11 document. The background is wider. How is this if the original document was only 8 1/2” wide? How or why would a document ‘grow’ by being scanned only then to have clipping mask applied to ‘shrink’ it back to 8 1/2?
It would appear the Xerox machine is filled with magic to do such things.
This is common to all scanners. No two scans are identical, even if made without moving anything, because the scanning process is quite sensitive to environmental conditions. It’s not the software algorithm, but the actual scanning that causes the difference.
What makes the huge difference between how the White House’s copier handled the long-form and how it handled the short-form, which the whitehouse.gov page clearly says is taken from the snopes page?
The White House image of the snopes image is at http://www.whitehouse.gov/sites/default/files/rss_viewer/birth-certificate.pdf
The snopes image that it cites is at http://msgboard.snopes.com/politics/graphics/birth.jpg
That printout showed through from underneath the long-form when the long-form was copied for the press. You can see a photo of that print-out for the press at http://theobamafile.websitetoolbox.com/post?id=5859778 (This phenomenon itself is a big problem because security paper is supposed to block anything behind it from appearing. And even if a scanner could be set to scan so deeply/sensitively that it could catch anything beneath it, it would surely have also detected the cross-hatches on the security paper.)
The snopes image also has the security background. If they printed that out on a computer printer it would still have the security background. If they then scanned the copy using the Xerox 7655 to create a PDF version of it and then sent it to a Mac computer and previewed it to correct the landscape orientation, it should have gone through the same kind of processing by the 7655 as the long-form did.
So.... How and why did they get rid of the security background on both the short-form copy for the press and the long-form copy for the press?
And how does the 7655’s performance on those 2 copies compare with its performance on the long-form? What layers and movable portions exist in those copies? They were presumably handled with the same settings and workflow as the long-form, since they are pdf’s that were sent to a mac and saved.
What 6’s and 8’s were substituted in all of these documents that went through the same workflow?
Most of this ‘Mixed Raster’ stuff is pure FUD.
Scanners scan. The original is always just a simple non-layered bit map. Always. Everything else some sort of software post processing. Always.
Now the device or computer program may use the bit-map to create another file in another format (such as Adobe pdf) but it is after the original image is captured. If this is done the original bit map may be discarded and never seen. Thus you are left with only the resulting converted file - a pdf, jpg, bmp or even text (if the desired output was just OCR output).
Adobe Standard and Adobe Professional have most of the ‘enhancing’ functions one would need. A native scan ‘printed’ in Adobe pdf will still contain a single image - a copy of the original bit map. If you attempt select sections of the document using the mouse you will get a blue box on a pure ‘printed’ copy that has no OCR yet applied. That is because the file has understanding of any text in the image - it is still treating it like a single monolithic image with no layers or OCR table.
In Adobe Standard (or Professional) you can have Adobe create an OCR table. So then you can highlight TEXT with your mouse and cut/copy and paste TEXT to another document. Adobe Standard/Professional will add the OCR table with a few clicks from the menu. Does it change the appearance of the document - hell no. The physical appearance of the document looks exactly the same. But the software ‘scanned’ the image looking for things it recognizes as text characters and creates a table inside the file. If you then highlight with your mouse the text characters will get highlighted in blue, not just a single box. The Obama LFBC has no OCR table in it. Even if did it would (or should not) not change the appearance of the image. When the OCR excuse was first used by this same crowd it was laughable.
Can you ‘optimize’ with Adobe Standard/Professional? Yes. And when it is done the text will change to make the new file smaller (optimized). And when it does the pixels are larger (like they are in Obama’s LFBC) and they have less variance in color and are mostly black (like Obama’s) also. What you do not get in any fashion is the stupid halos and the other variances in pixelation. The halos appear in manufactured images, not ones that have been reformatted for standard purposes.
Another non-technical red flag is the Nordyke-like shadowed left bend. That should not be there. Its in the Nordykes certified copies. But for good reason - their COPIES were made in 1965! Well before the originals were scanned into a digital (likely optical WORM) library. Images stored in the digital library are SCANNED FLAT. Except for a very slight bend the images printed on security paper are flat in their appearance. That Nordyke-like left bend is a phenomena of a copy being made while the document is still in a bound volume. To scan them - they unbind them and scan them flat. There has yet to be another example of a LFBC image printed on security paper with a transparent background that looks like Obama’s LFBC. In this regard is truly a only-one-on-this-earth ‘document’.
We will not even get into centered fields with a typewriter. If you ever took a typing class or really had to use a typewriter (manual or electric) you would know what a royal pain centered fields are - ROYAL PAIN! Computers can center things easily - real typewriters, no so much. This is why every LFBC has tabbed fields. Typewriters do do tabs well. Centered fields with a typewriter - yeah right. Stupid 20-somethings or even 30-somethings at work replicating something they only saw in a museum.
This thing is a mess. A complete mess - every way one looks at it.
the security line discussion is a red herring
there should be no way to remove the background in the first place, if it was scanned from one image and printed ONTO security paper.
but if you did use Photoshop and remove the black ink from the green background then there would be no way for the program to know what was behind the ink- it would not automagically fill in the security paper lines.
What is Papit doing to give it different looking security lines? Using photoshop? Or could be that it is Dr. Conspiracy is tinkering with the image he was sent in order to make Papit look deceptive or not experienced?
The White House image of the snopes image is at http://www.whitehouse.gov/sites/default/files/rss_viewer/birth-certificate.pdf
This source of this image is actually just a printout from MS Explorer. The header and footer are standard for a print out from Explorer. Someone printed this originally with a browser.
Here is the substantive part of NBC’s reply to bluecat6:
http://nativeborncitizen.wordpress.com/2013/08/16/bluecat6/
Bluecat6
Posted on August 16, 2013 by NBC
Bluecat6: Guthries is even bigger than the image after the clipping mask is removed. And the two ‘dots’ on the right side of the backgroup with the clipping mask do not show up in her ‘photo’.
NBC: You have to be careful because you are now looking at a document which is scaled based on the distance of the camera. You can calibrate this by measuring a known distance on the PDF or jpeg such as the length of one of the boxes on the LFBC.
Bluecat6: All three are physically different and can easily be verified by the cross hatch pattern.
NBC: That is correct, remember that the Xerox removed 0.12” of borders, which explains why the photographs show more information.
Bluecat6: But clipping mask was applied to create an 8 1/2 by 11 document. The background is wider. How is this if the original document was only 8 1/2” wide? How or why would a document ‘grow’ by being scanned only then to have clipping mask applied to ‘shrink’ it back to 8 1/2?
NBC: It was not the clipping mask, but the edge erase. This has puzzled me a bit as well but if you carefully count the weave marks, you will see that it starts to make sense. Note that the clipping mask does not remove data, it just hides it.
I made the blog.
Explain the left bend shadow. Why is Obama’s the only Hawaii LFBC ever printed on security paper with a transparent background that has the shadowed left-bend similar to the Nordykes that was created with completely different technolgy in 1965.
24 hours to find a few examples. The clock is ticking.
Substantive?
‘edge erase’?
Really?
But the PRESENTED part is 8 1/2. That means the image cut out is LARGER than 8 1/2” if it came from a real document.
Why edge erase or clip mask anything? This is now reverse engineering the explanation to fit the problem.
Lets see some other documents with the shadow left-bend printed on security paper. If no examples in 24 hours - then I call BS.
Oh,
And they better have centered manually typed fields or its double BS.
The photo of the printout for the press is of a black-and-white copy that was the first page in a 4-page presspacket. Since it was printed on regular paper, the second page is showing through. The photographer to photos of all of the pages of the packet. Page 3 shows through on the photo of page 2, and page 4 shows though on the photo of page 3.
“Explain the left bend shadow. Why is Obama’s the only Hawaii LFBC ever printed on security paper with a transparent background that has the shadowed left-bend similar to the Nordykes that was created with completely different technolgy in 1965.”
The original birth record was bound into a book with a total of 500 records. The book was opened to the page containing Obama’s birth record, placed on the scanner glass, blank security paper was placed in the paper tray, and the Copy button was pressed. HDOH has confirmed that they still have the original bound hardcopy (they announced that they had moved it into a secure case within their already secure vault for added security). If it looks like it was done using the same technology as 1965, well, that’s because photocopies are basically the same. A bit crisper, and with color capabilities and the ability to save in a digital format, but it’s the same basic technology.
The PRESENTED part includes the edge erase. The visible green basket-weave background is only 8.17”x10.5”. The white is the clipping mask (which is larger than the edge erase, so the edge erase is not visible). The background jpeg is slightly larger than 8.5x11, but only a few pixels (this is likely due to jpeg compression using 8x8 blocks). There is no evidence for an original larger than 8.5x11
How did they get rid of the security background?
You mean like the Nordyke’s?
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.
