Yep. And Hawaii's NOT doing something else that would really help in what is all our duty as citizens and officials under the Constitution.
That is, Hawaii should release all pertinent records pertaining to the Mr. Obama. Otherwise they have joined the stonewalling.
AJ is correct on several of his assumptions such as the OpenDNA images came after the KOS image and that there was a change in the COLB certificates between the Decosta and the KOS versions. However I have been using a half dozen separate Hawaiian COLBs issued over a 6 year period all the way through 2008 - some hi-res digital scans and some sitting on my desk (real ones) as comparison specimens. And once again to summarize - none of them match the security border of the KOS image. Unlike AJ I just happened to have been lucky enough to have had more than just the Decosta image to work with so that I am able to see the progression of the document modifications through time. (I also called Hawaii and talked to them about it) But from AJ's point of view his observation would be correct having not seen a 2006, other 2007's, or 2008 COLBs. I also realize that there has been some confusion and I am not basing my analysis on the border itself but on the analysis of the actual overall modifications made to the KOS image. The border just happened to be what caught my attention in the first place. Long story short the KOS image is made up of several discrete layers (changes) that can be viewed by using advanced image analysis techniques. [geek alert] Any recent changes made to a JPEG are for lack of a better word "recorded" by the individual 8x8 pixel cells that make up the JPEG. By measuring the differences in error rates and performing wavelet and PCA analysis you can actually show the individual modifications. Mix that with basic observational analysis and metadata extraction and you can form a more complete picture. In the case at hand the KOS COLB was not just simply scanned into Photoshop and a box placed over the certificate number and then exported for the web. The original scan has been saved multiple times, the text and the borders have both been modified, and the background image (the paper) may be the only thing still "real" in the image. Unfortunately there is no way to know what the original state of the document was (you can not bring things back from the dead at least I can’t) but it is not in the same state as the original scan. The best I can do is show you what changes were made with an estimate as to order in which they were modified. Why they were made and who made them is not my concern.While AJ and I may poke fun at each other it is not in a mean spirited way. And on a serious note I would like to get AJ’s take on the analysis techniques I have been using. A whitepaper presentation by Dr. Neal Krawtz on which some of my methods and algorithms are based on can be found here http://www.blackhat.com/presentations/bh-usa-07/Krawetz/Whitepaper/bh-usa-07-krawetz-WP.pdf and one by Jessica Fridrich et al http://www.ws.binghamton.edu/fridrich/Research/copymove.pdf
So everyone keep your shirts on - I am writing up the full report and it will be done as soon as I have the time with all of the explicit details and lots of geeky stuff tossed in for good measure. I still need to finish capturing screenshots, resize them a bit, and then put together a section of the report so you can all do the same experiments at home using some freely available analysis tools.
But on a side note if AJ or anyone else has any questions for me feel free to post them. I will answer anything you ask to the best of my abilities but you might want to wait a few more days so you can read the entire report - so far just explaining the methodology and the science behind them has put it over a dozen typed pages (I obviously also need to edit them down a bit). It may have to be posted in parts at this rate. P.S. - For giggles I also used the same analysis method against McCains posted COLB...it is not modified (are you shocked?) and tossed it into the report for good measure. I also added in a few known modified images so you can see how the process works.
Posted by: techdude | Monday, July 14, 2008 at 12:58 AM
“Hawaii should release all pertinent records pertaining to the Mr. Obama.”
I don’t believe Hawaii can legally do that, I am not sure.
You can certainly try to put in a FOIA request.
Logically, the burden is on Obama, he’s the candidate.
As far as i can tell, the core of the disagreement is this:
“What Techdude took as a really, really bad forgery I had always concluded represented an updated background and paper for the Hawaii COLB. It never occurred to me any serious adult would confuse two clearly different documents, separated by 5 years, as being some kind of forgery.”
Either the Obama COLB got the background ‘wrong’ and is a forgery, or the Obama COLB has a background which matches the real form output for those who made the request in June 2007, thus validating its varacity as a scan of a real COLB.
The DeCosta 2002 form doesnt conclusively prove a forgery, as the background may have changed. Hawaii freepers could request COLB copies *today* and thereby shed light on the format.
They CAN'T. They can prosecute for election fraud and falsification of official documents, and the judge may or may not decide to release the true documents.
But due to the right to privacy they can't release his documents without his permission or a court order.
Thank God.