First a disclaimer. I am not a "Birther." However, I was curious when President Obama produced the "long form" of his birth certificate. I expected that skeptics would be looking for any possible flaws in the document, so I thought I would want to look at it. Fortunately, the White House released a high quality PDF version of the document. I expected the document to be a flat file PDF, created by a simple scan of the original document. As it turns out, it is anything but what I had expected. In fact, the PDF version published online was highly digitally edited and assembled in "pieces."
Instead of being a flat image, the PDF file published by the White House consisted of multiple images that had been overlaid over each other. It was immediately clear to me that somebody had scanned the image, knocked out the text (but leaving light gray ghost around each part of the text). This fact was easily confirmed through a PDF editing program (PDF Converter Professional 8.0). The image right is a screen shot of the PDF editing program, with all images within the PDF file selected (surrounded by blue rectangles). Clicking on the image to the right will open the full-sized version in another window, which will make it much more easy to view.
The PDF file consists of nine individual images. A close-up of the central part of the above image can be seen at the right. As can be seen, several "text" images have been overlaid on the original image, including parts of the stamped dates. However, all the small overlaid images are much lighter in color than the other scanned text images. A poor job was done trying to match the brightness of those images when the scans were reassembled.
The background (main image) consists of the original scan, which was a semi transparent image that was overlaid on top of the green patterned background. For some reason, much of the black text was removed from this image, leaving a white/gray ghost around all the words. The lines that compose the form do not have these ghost around them, since they were part of the original scan. Handwritten (as opposed to typed) random numbers and x's appear as light gray images in this layer. However, the background is otherwise quite clean for a scan of such an old document. It is quite likely that stray pixels were cleaned up and removed from this layer.
Main overlaid text image
The text that was knocked out of the original scan was overlaid on top of the background as a separate image. It appears that the reason for this procedure was to darken the text, since the text in the text overlay is much darker than corresponding adjacent text from the background image. Again, a poor job was done trying to match the brightness of the background scan. All of the text represented in the overlaid text image was surrounded by white/gray ghosts on the background image, showing it had been removed, darkened and put back.
Other text overlay images
Several small texts overlay images were separately included in the document. It might be expected that the date of the request and the state registrars stamp might appear as separate images, since they were not part of the original birth certificate. However, part of the date accepted stamps ("AUG - 8 196") were included as separate images. In both instances, the "1" from "1961" was included in the main text overlay, so was not part of these images. The other separate image was the "Non" from item #17, occupation of the mother. As in the other date images, the last letter ("e" from "None") was included in the main image overlay and was not part of the small image overlay. As mentioned previously, all these images are less dark than the corresponding parts from the main overlaid text.
Why was it done this way?
The Obama birth certificate PDF document is amazingly readable, probably in part because of the extensive image editing that was done to enhance the parts that were probably not as readable in the original scan. It appears that the original scanned image was highly processed in Adobe Photoshop or similar image editing program. Many parts of the original form (lines and some of the text and signatures) apparently scanned well and were included in the assembled background image, which was overlaid on the green patterned background. It is obvious that the green background was never part of the original scan. Why was it included at all? Who knows. Other parts of the text were probably not as dark, and were selected and independently enhanced for contrast. Apparently, the two stamped dates and "Non" were even less dark and had to be enhanced separately. You may have noticed a have not used any of the "f-words" (forgery, fake, etc.). It seems to be just a case of somebody with poor image editing skills attempting to clean up the original birth certificate scan. However, all the obvious photo editing is going to get the Birthers in an uproar about how fake the birth certificate is. Who did all this editing? My guess is that it was done by the office of the Hawaii State Registrar. One would think that if it were done by somebody in the White House, they would have gotten somebody with better photo editing skills to do it. The only clues we have are that the document was created on a Macintosh computer running OS X 10.6.7 running their "Quartz" graphic rendering library, which printed the graphics in layers. If had been printed on a Windows-based computer, no layers would have been created in the PDF document (I tried it). If you want to look at all the layers using Adobe Photoshop, please download the assembled document below:
It appears that the birth certificate released by the White House is an authentic, though highly edited facsimile of the original. All of the enhanced text is surrounded by white/gray ghosts on the background image, which would have been left by somebody who was an amateur at editing scanned images. Since these background ghosts correspond exactly with the enhanced text, it would be unlikely that the document was composed or edited for content.
A new article has appeared that points out signs of definite digital copying of certain numbers and boxes that would never be produced in a real scan of a document. The fact that these images are the ones which would be expected to be replaced in a "Photoshopped" forgery using the Nordyke twins birth certificates as a model is very troublesome:
Last Modified May 3, 2011