Posted on 08/07/2013 6:29:11 PM PDT by Seizethecarp
And BTW, why use a machine that used OCR when you want a mirror image of the document a bit by bit copy. There was no need to determine what character is what. If it is a “6” or is an “8” ? You would not. You want a picture a photograph like image.
These clowns are blowing smoke up our butts.
What artifacts do you mean, that aren’t on the AP scan? By AP scan you mean the Applewhite image?
thecodont asks
“I am not an expert on this, but would Xerox be using a grayscale (8 bit) to black and white (2-bit) conversion for scans of these text images? Grayscale, of course, provides more information for further processing. You would have to have multiple scans of 8-bit to 2-bit of the same test image and then examine the results under a magnifying glass.”
Here is NBC’s response:
Nope, a monochrome bitmap requires 1 bit (on/off) with 2 bits you can create 4 colors (00, 01, 10, 11). As to looking under a magnifying glass, remember that these are digital documents so we can use better tools. The reason to use B&W is that the image can be stored in 8 times as little information. It’s all about compression here.
Good questions. Finding the identical characters is not much fun and I wish I had a tool to do so. But alas.
http://nativeborncitizen.wordpress.com/2013/08/10/butterdezillion-some-good-questions/#more-46824
“...why use a machine that used OCR when you want a mirror image of the document a bit by bit copy. There was no need to determine what character is what. If it is a “6” or is an “8” ? You would not. You want a picture a photograph like image.”
NBC is not claiming that OCR recognized a “6” or an “8” as a number. NBC contends that the Xerox copier uses an international JPEG standard software called “Mixed Raster Compression” incorporating an algorithm called “image segmentation” to identify borders and shapes. If a shape is repeated closely, then the first one is copied rather then saving a second or subsequent shape that is only slightly different.
http://en.wikipedia.org/wiki/Mixed_raster_content
“Mixed raster content, or MRC, is a method for compressing images that contain both binary text and continuous-tone components, using image segmentation methods to improve the level of compression and the quality of the rendered image.[1] By separating the image into components with different levels of compressability, the most efficient and accurate compression algorithms for each type can be used.”
http://en.wikipedia.org/wiki/Segmentation_(image_processing)
“Compression based methods postulate that the optimal segmentation is the one that minimizes, over all possible segmentations, the coding length of the data.[5][6] The connection between these two concepts is that segmentation tries to find patterns in an image and any regularity in the image can be used to compress it.”
So when the compressed JPEG is opened in Preview on a Mac and then saved as a pdf file, instead of four similar lower case letter “e” images appearing in the LFBC, there is only one “e” repeated four times, IIRC. If you read the NBC blog there is a lot of explanation of this.
Yes, the Applewhite image.
NBC has an extensive analysis and back and forth with a doubter here:
http://nativeborncitizen.wordpress.com/2013/07/29/following-the-applewhite-document/
“Now check out the Metadata. The caption indicates that it is a photo (AP Photo J. Scott Applewhite) and the date/time shows that the photo was captured (EXIF) Date Time Original: Apr 27, 2011 8:53:21 AM. So let’s look at a plausible workflow: Applewhite used Raw Camera Capture to import the RAW photograph into Photoshop and saved it as a JPEG. The resulting JPEG was sent to the DC office or alternatively, the conversion was done at the DC office based on the raw data. Note how the Creator is Adobe Acrobat 8.26 and the Produced Adobe Acrobat 8.26 Image Conversion Plug-in. Someone created a PDF from a converted image.”
I can’t find it at the moment but, IIRC, NBC claimed that the replicated lowercase “e” for example in the WH pdf is NOT present in the AP/Applewhite image.
If I am following NBC’s argument correctly, the Applewhite image did NOT use the same Mixed Raster Compression to create the JPEG employing the image segmentation algorithm that was used by the WH Xerox copier.
MRC as an explanation has been around since the LFBC PDF was first release.
In the comments of this ORYR article from 2011
http://obamareleaseyourrecords.blogspot.com/2011/04/associated-press-involved-in-new-obama.html
@Anonymous
The artifacts you see are a result of PDF compression...in this case a kind of Mixed raster compression alrogithm was used. You guys are silly...
April 30, 2011 at 9:35 AM
Also when NBC talks about the JPEG, I believe the reference is to the green background layer. That layer is a JPEG and the other layers are bit maps.
The PDF was created on the Xerox by the Xerox. Then e-mailed to a user who opened it on a Mac in Preview.
Why would he base it on anything else?
Thanks for the Pings, as always.
I’m wasting way too much time, looking at PDF’s posted at whitehouse.gov. The tax return that was posted to the whitehouse.gov website in April of 2011 is the only time I can ever find Xerox mentioned. All the other times it’s either Mac OS or a verson of Adobe for the PDF Producer and/or Application.
The long-form and short-form BC’s that are posted on the internet are different from all the other documents in that the PDF was produced on a Mac.
And the 2011-posted tax return that NBC cites is different from all other tax returns posted, in that the “Properties” section mentions the Xerox for both the PDF Producer and the Application, rather than an Adobe version for PDF Producer and/or Application.
That’s looking at all the apparently-scanned documents I can find posted on whitehouse.gov from 2009 to 2013; the one in 2011 is the anomaly. If the White House uses the Xerox WorkCentre, then the scanned documents should all show that for the PDF Producer and Application - unless something was done differently so that the properties show different PDF creators. IOW, if the tax return scan showing Xerox WorkCentre as the PDF producer is the way the properties appear on a normal Xerox WorkCentre scan, then what we know is that the long-form, short-form, and all the other tax returns were done some way besides the normal scan.
But it sounds like he is saying that it is very temperamental. For instance, it might recognize two typewriter a’s as being close enough to just copy the first one where the 2nd one goes - but wouldn’t recognize 2 boxes with just a pixel or two off. And the German researcher found that it would substitute 8’s and 6’s. If an 8 is close enough to a 6 but a box with a pixel or two off isn’t close enough, something seems screwy to me.
Mr. K, you said at one point that these claims are baloney, noting that you have programmed imaging software. Do you have any insights as to how the compression could be insensitive enough to mistake an 8 for a 6 and yet not recognize every 6 as being the same?
Also, anybody with an engineering, architecture, or construction background, tell me the likelihood that 8’s and 6’s have been switched around on Xerox copies of blueprints for the past 9 years and nobody has noticed the problem in all that time.
“The artifacts you see are a result of PDF compression...in this case a kind of Mixed raster compression alrogithm was used.”
Talk is cheap. It took two years before NBC published a pdf claiming to actually reproduce the many “artifacts”...
Does the Applewhite image have the material that was hidden by the clipping mask?
“Why would he base it on anything else?”
I don’t see Hayes surviving a Daubert challenge to his qualification an expert on Xerox JPEG compression is what I meant from what is in his CV.
“Does the Applewhite image have the material that was hidden by the clipping mask?”
YES, and NBC claims to have explained why in great detail in posts on his blog.
http://nativeborncitizen.wordpress.com/2013/08/09/xerox-7655-clipping-path-and-objects/
“The following three images show the ‘raw’ Xerox file, which has no clipping path at the top level and the Preview saved version which shows that Preview added a clipping path.”
The Applewhite image was not processed through Preview so it doesn’t have the clipping mask. NBC claims to have demonstrated that it is Preview that added the clipping mask.
Why would they scan it on the Xerox and not save it until it was on a Mac? They didn’t do that with the tax return NBC cites. Of course, that tax return is the only one that mentions Xerox WorkCentre. If the Xerox does all scans in landscape orientation, then wouldn’t the tax form be sideways unless it was first previewed, rotated, and then saved - yielding a PDF producer other than the Xerox?
I knew something was screwy with the WH long-form file when I tried C&P-ing it into a Word document and it showed the haloes. I wanted to test whether the same thing happened with other documents scanned and posted at whitehouse.gov but my computers are giving me trouble again.
Just so you know, Zullo has mentioned that the security background doesn’t match that used at the HDOH at the time, so his emphatic claim that the White House image is absolutely known to be a forgery is not just based on the artifacts and/or computer analyses.
The Applewhite image has the two dots at the right edge?
Also, how did the White House clean up the file a couple hours after initially posting it, when they realized that it wasn’t flattened? Is the 2011 tax return similarly unflattened? The 2009 tax return was unflattened, but that one had the SSN redacted so it had to be manipulated, yielding the layers that needed to be flattened. The 2011 one must have had the SSN redacted as well, meaning that it had to have layers as well. Would it have to be previewed and saved on a different computer to flatten the file?
This paragraph pretty much sums it up. This falls under the "too good to be true" category. Follow the timing by overcomplicating the explanation, which is just intended to confuse enough people to create doubt about Zullo's investigation. Even with the multiple blogposts worth of "explanations," there's a problem if there was ANY kind of manipulation after the original scan. And there's a second problem when the layers and manipulations can be EASILY explained by the process of creating a PDF from a digitally fabricated documented, such as through InDesign and then converted to PDF. Nothing offered in the new explanations can rule that out.
Understood. I'm saying that I don't think the CCP would employ Hayes in a fashion that would jeopardize his Daubert-worthy credentials, and I am certain that Hayes himself would start dropping qualifiers if he thought he were venturing into areas of expertise not his own, particularly when he's been commissioned to give his expert opinion on a particular item, and especially one as weighty, implications-wise, as this one.
I feel no churn over this "Native Born Citizen" blog's claims. Seems to me more like they and other Obots are $hitting egg rolls over Hayes' report and the marked progress being made by the CCP in the last few months.
I'm, in Mickey D's parlance, Lovin' It.
Some of Binden’s returns also show that they were scanned on a Xerox 7655.
“Why would they scan it on the Xerox and not save it until it was on a Mac?”
The Xerox WorkCentre emailed the PDF directly to the users in-box. That person opened it on a Mac, but it could also have been opened on a windows based computer.
“If the Xerox does all scans in landscape orientation, then wouldn’t the tax form be sideways unless it was first previewed, rotated, and then saved - yielding a PDF producer other than the Xerox?”
The tax forms are in landscape when you open them in Adobe Illustrator. But when you view them in Adobe Photoshop, Acrobat Reader or Preview they are in portrait mode. The software seems to recognize the orientation.
As to the timing. the tax returns are not just available at the White House website, they are also on other sites like the Tax History Project. Here is the wayback machine’s view of the THP site from April 20, 2011.
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.