MS – Orly v Democrat – Henry Blake Affidavit – Part 1

In one of the pending cases, a person named Henry Blake, submitted an affidavit in which he compared various versions of President Obama’s long form birth certificates with the White House version

  1. 05/04/2012 15 MOTION for Judgment on the Pleadings by Democrat Party of Mississippi (Attachments: # 1 Exhibit LFBC from White House, # 2 Exhibit COLB from Campaign, # 3 Exhibit DOH Verification re White House BC, # 4 Exhibit Hawaii Gov April 27 2011 News Release, # 5 Exhibit DOH White House Correspondence, # 6 Exhibit DOH 08-93 News Release, # 7 Exhibit DOH 09-063 News Release, # 8 Exhibit CDC Report re Birth Certificate History)(Begley, Samuel) (Entered: 05/04/2012) [.pdf (page 8)]
  2. 06/06/2012 35 MOTION to Supplement Counsel for MDEC’s Response 30 in Opposition to Plaintiff Taitz’s Motion for Sanctions 25 re 30 Response to Motion, by Democrat Party of Mississippi (Attachments: # 1 Exhibit MDEC Counsel Request to HI DOH for Verification of President Obama’s Hawaiian Birth Cert, # 2 Exhibit Hawaii DOH Verification of President Obama’s Hawaiian Birth-Issued May 31 2012)(Begley, Samuel) (Entered: 06/06/2012) [.pdf (page 11)]
  3. 15-1.pdf ( 10513207156.pdf ) 05/04/2012 15 MOTION for Judgment on the Pleadings by Democrat Party of Mississippi (Attachments: # 1 Exhibit LFBC from White House [.pdf(page 2)]
  4. 35-1.pdf ( 10513240131.pdf ) 06/06/2012 35 MOTION to Supplement Counsel for MDEC’s Response 30 in Opposition to Plaintiff Taitz’s Motion for Sanctions 25 re 30 Response to Motion, by Democrat Party of Mississippi (Attachments: # 1 Exhibit MDEC Counsel Request to HI DOH for Verification of President Obamas Hawaiian Birth Cert [.pdf (page 4)]

I am not sure why he looked at two identical filings as 1/3 and 2/4 are the same documents. Not surprisingly he observes that

Page 8 of the first document above is nearly identical to the white House LFCOLB PDF image file and is identical to page 2 of the third document. Page 11 of the second above document is identical to page 4 of the fourth document. For convenience I will denote the two copies that I recently examined as (page 2/8) and as (page 4/11).

He first “observes” that the:

The (page 2/8) image copy is otherwise identical to the white House image except for the “Case Label” added at the top margin.

Which is of course trivially true since the documenting system used by the court will add these labels. Most likely they are added as a simple ‘stamp’ object and thus remain selectable.

The two case labels at the top edge of the (page 2/8) image copy suggest that the (page 2/8) image was altered to make the (page 4/11) image copy. Other findings reported herein for the first time indicate otherwise.

Well, yes, the second document show the first document photocopied from the first document and uploaded, hence it receives a second documenting stamp.

The LFBC attached in documents 1 and 3 matches the document provided on the Whitehouse site perfectly as they are identical. I have extracted the layers in both documents and they are the same.

PDF indeed shows a Form that includes the watermark on each page for ‘Friend of TheFogBow.com”

The black text is from the original document found at the Whitehouse

<<
/Subtype /Watermark /Type /Pagination >>
BDC  q 1.6324615 2.2469025 -2.2469025 1.6324615 11.9998779 103.6101379 cm /GS0 gs /Fm0 Do Q

EMC  
q 18 14.4     576 763.2 re W n q 0 -792.9600067 612.4799957 0 -0.2400055 792.4799957 cm /Im0 Do 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 
q /CS1 cs 0.106  0.176   0.122  scn   0 -348.9600067 436.5599976 0 89.2799988 581.2799988 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 /CS1 cs 0.345   0.392  0.353  scn  0 -47.7599945 186.7200012 0 304.5599976 108.9600067 cm /Im2 Do Q 
          0.34510 0.3922 0.3529 sc q 0 -47.76      186.72      0 304.56      108.96      cm /Im3 Do Q
q /CS1 cs 0.302 0.345   0.322  scn  0 -10.0800018 65.7599945 0 170.1600037 89.7599945 cm /Im3 Do Q 
          0.302 0.34510 0.3216 sc q 0 -10.08      65.76      0  170.16     89.76      cm /Im4 Do Q
q /CS1 cs 0.255  0.337  0.263  scn  0 -29.5200043 54.7200012 0 440.3999939 274.0800018 cm /Im4 Do Q 
          0.2549 0.3373 0.2627 sc q 0 -29.52      54.72      0 440.4       274.08      cm /Im5 Do Q
q /CS1 cs 0.341  0.435  0.341  scn  0 -11.2799988 51.8399963 0 103.4400024 254.8800049 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
q /CS1 cs 0.255  0.337  0.263  scn  0 -8.1600037 16.8000031 0 349.6799927 322.0800018 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
q /CS1 cs 0.941  0.973  0.922  scn  0 -58.3200073 52.0800018 0 176.1600037 185.7599945 cm /Im7 Do Q 
          0.9412 0.9725 0.9216 sc q 0 -58.32      52.08      0 176.16 1    85.76       cm /Im8 Do Q
q /CS2 cs 0.965  scn  0 -31.6799927 34.0800018 0 251.7599945 784.8000031 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'
Q BT 0 0 1 rg /T1_0 12 Tf 96.24 767 Td 
(Case 3:12-cv-00280-HTW-LRA   Document 15-1    Filed 05/04/12   Page 2 of\\\r 2)Tj ET '

The labels at the top of the pages is a simple text string rendered. It is clear that the documents are closely related and by doing a binary diff, I have found that the image files are identical.

So nothing mysterious here.

As to the second and fourth document. I believe that we can unravel the minor mysteries here as well by observing that OCR was turned on. It helps being able to look at the actual code rather than using Illustrator.

AFFIDAVIT

I Henry Blake, do hereby state under penalty of perjury that the following facts are true:

I. I am over the age of 18 years old and am a resident of Tennessee. The information contained in this affidavit is based upon my own personal knowledge and, if called as a witness, I could testify competently thereto. I attained higher education as follows: BS Mechanical Engineering, MS & PhD in Engineering Mechanics. I have over 40 years experience with a variety of computers including: Personal Computers, Work Stations, Mainframes and Super Computers.

II. I have general knowledge of foregoing legal filings in the Federal District Court Case 3:12-CV-00280 HTW-LRA and have specific knowledge of the digital images of copies of President Obama’s purported Long-Form Certificate of Live Birth (LFCOLB) appearing within the case filing documents.

III. I have examined the two image copies of the purported LFCOLB which exist as digital electronic page documents, namely

05/04/2012 15 MOTION for Judgment on the Pleadings by Democrat Party of Mississippi
(Attachments: # 1 Exhibit LFBC from White House, # 2 Exhibit COLB from
Campaign, # 3 Exhibit DOH Verification re White House BC, # 4 Exhibit Hawaii
Gov April 27 2011 News Release, # 5 Exhibit DOH White House Correspondence,
# 6 Exhibit DOH 08-93 News Release, # 7 Exhibit DOH 09-063 News Release, #
8 Exhibit CDC Report re Birth Certificate History)(Begley, Samuel) (Entered:
05/04/2012) [.pdf (page 8)]

and

06/06/2012 35 MOTION to Supplement Counsel for MDEC’s Response 30 in Opposition to Plaintiff Taitz’s Motion for Sanctions 25 re 30 Response to Motion, by Democrat Party of Mississippi (Attachments: # 1 Exhibit MDEC Counsel Request to HI DOH for Verification of President Obama’s Hawaiian Birth Cert, # 2 Exhibit Hawaii DOH Verification of President Obama’s Hawaiian Birth-Issued May 31 2012)(Begley, Samuel) (Entered: 06/06/2012) [.pdf (page 11)]

see also court documents

15-1.pdf ( 10513207156.pdf ) 05/04/2012 15 MOTION for Judgment on the Pleadings by Democrat Party of Mississippi (Attachments: # 1 Exhibit LFBC from White House [.pdf(page 2)]

and

35-1.pdf ( 10513240131.pdf ) 06/06/2012 35 MOTION to Supplement Counsel for MDEC’s Response 30 in Opposition to Plaintiff Taitz’s Motion for Sanctions 25 re 30 Response to Motion, by Democrat Party of Mississippi (Attachments: # 1 Exhibit MDEC Counsel Request to HI DOH for Verification of President Obamas Hawaiian Birth Cert [.pdf (page 4)]

and which were filed by the Mississippi Democratic Executive Committee (MDEC), as well as the original PDF image of President Obama’s LFCOLB, which was posted on the White House web site on Apr. 27, 2011,
http://www.whitehouse.gov/sites/default/files/rss_viewer/birth-certificate-long-form.pdf
and have found significant alterations were made to the original White House LFCOLB PDF image file.

Specifically the copy of the LFCOLB appearing in the second and fourth documents above differ significantly from the copy appearing in the first and third documents and from the original copy posted on the White House website.

Page 8 of the first document above is nearly identical to the white House LFCOLB PDF image file and is identical to page 2 of the third document. Page 11 of the second above document is identical to page 4 of the fourth document. For convenience I will denote the two copies that I recently examined as (page 2/8) and as (page 4/11).

IV. The alterations made to the White House LFCOLB PDF image to create the (page 2/8) image copy are incidental.

1. The (page 2/8) image copy is otherwise identical to the white House image except for the “Case Label” added at the top margin.
2. Only the Case label is selectable by mouse and cursor on (page 2/8)
3. The two case labels at the top edge of the (page 2/8) image copy suggest that the (page 2/8) image was altered to make the (page 4/11) image copy. Other findings reported herein for the first time indicate otherwise.

V. I found that the alterations to the (page 2/8) image copy to create the (page 4/11) image
copy are extensive as follows:
1. A second Case Label was added
2. The (page 2/8) image was flattened, rasterized and green color was added
3. The color of all text was changed from dark Green-Black to light Green-Black
4. Numerous form lines were repaired or replaced entirely
5. The basket-weave background was softened and touched up
6. Specific words were made selectable by mouse and cursor in Adobe Reader
7. Hidden editing was applied (see 8.- 10. below and Exhibit 1):
8. Twelve Line Objects were added
9. Two Broad-line Strikeouts were applied, one to the words “Kapiolani Maternity & Gynecological”, and the other to the word “August”
10. Seven Rectangular Black Redaction Box Objects were added which altogether cover
nearly all of the typed text. (See Exhibit 1)
11. Two patches were applied to the green basket weave background at the bottom corners of the page. (See Exhibit 2).
12. The two shortest, hidden line segments of the 12 added line segments were associated with a vertical line appearing on the (page 4/11) LFCOLB image. (See Exhibit 3).
13. The two White-splotch regions existing in the WH LFCOLB PDF image and the (page 2/8) LFCOLB PDF image are missing from the flattened (page 4/11) LFCOLB PDF image. (See Exhibit 4).
14. Two of the only four typed characters which are located on the Green basket-weave background layer of the (page 2/8) LFCOLB PDF image were compared to the same two characters on the (page 4/11) image and were found to be different in appearance. These characters are two of the only four that are 150 PPI resolution in both images. All of the remaining characters are 300 PPI resolution on the (page 2/8) image and 150 PPI resolution on the (page 4/11) image. (See Exhibit 5)
15. All of the text fonts of the (page 4/11) LFCOLB PDF exhibit pincushion distortion. The text fonts of (page 2/8) LFCOLB PDF image and the WH LFCOLB PDF image are not distorted. Grid lines are everywhere straight in these two earlier LFCOLBs. (See Exhibit 6).
16. The faint seal impression that is visible on the (page 2/8) LFCOLB PDF image is missing from the (page 4/11) LFCOLB PDF image. (See Exhibit 7).

VI. The remaining typed, stamped (or form) words left unstruck or unredacted and appearing
on the hidden image of (page4/11) are:
1. Both Case Labels
2. State of Hawaii Certificate of Live Birth Department of Health
3. 61 10641 (file number 151 is redacted)
4. Barack Hussein Obama, II (“X” in single-birth box)
5. Honolulu (in two places)
6. 6085 Kalaianaole Highway
7. Labels for ten form-boxes
8. State Registrar’s date and signature stamps
VII. Words which can be selected on (page 4/11) by mouse and cursor in Adobe Reader are:
1. Both Case Labels
2. State of Hawaii Certificate of Live Birth Department of Health
2. 61 10641 (File number 151 is non-selectable)
3. Barack Hussein Obama, II
4. Honolulu
5. Kapiolani Maternity & Gynecological Hospital
6. 6085 Kalanianaole
7. Hussein Obama

8. University
9. State Registrar’s date and signature stamps.

General Findings

The two case labels applied to the top edge of the (page 4/11) LFCOLB PDF image imply that it was created from the pre-existing (page 2/8) PDF image. If true, then the (page 4/11) LFCOLB PDF image would be identical to the (page 2/8) LCOLB PDF image except for the second case label added to the top edge of the document. The (page 2/8) LFCOLB PDF already existed in PDF format as page two of Document 15-1.pdf. Thus there would be no need to create another Obama LFCOLB PDF image in the same law case.
Assuming all of this to be stipulated, then the task of creating the (page 4/11) LFCOLB PDF image could have been most easily accomplished as follows. To create the (page 4/11) LFCOLB PDF image requires only that page 2 of Document 15-1.pdf be extracted into a separate one-page PDF document and a second case label added above the first. This could all be accomplished in Adobe Acrobat which was available to whoever created Document 35-1.pdf. The four steps are:

1. Document 15-1.pdf is opened in Adobe Acrobat and then page 2 is extracted into a separate one-page PDF document.
2. The object containing the existing Document 15-1 case label is then selected and its color is changed from bright Blue to light Green.
3. The bright-Blue case label for Document 35-1 is then typed above the light Green case label of Document 15-1 within the margin of the (page 4/11) LFCOLB PDF image.
4. The resulting (page 4/11) LFCOLB one-page PDF image file is then merged with the three-page Tepper-to-Fuddy letter to create the Document 35-1.

The resulting (page 4/11) LFCOLB PDF image would then be identical to the (page 2/8) LFCOLB image except for the added second case label. The Blue second case label contains the Document number 35-1.

The (page 2/8) LFCOLB PDF image is identical to the WH LFCOLB image except for the first case label added to the top edge of the page. This first case label contains the document number 15-1.

It follows that, under this preferred work flow, the (page 4/11) PDF image would then be identical to the WH LFCOLB PDF image except for the two added cases labels.

Had this preferred work flow been applied to the creation of the (page 4/11) LFCOLB PDF image, then there would be a solid chain of evidence between the (page 2/8) LFCOLB PDF image and the (page 4/11) LFCOLB PDF image. Unfortunately, the collective findings reported herein indicate that this preferred work flow was not followed in this case.

The collective evidence indicates that the (page 4/11) LFCOLB PDF image which comprises page 4 of the court Document 35-1.pdf is at best a scanned image of some other Obama LFCOLB document. At worst it is an entirely new Obama LFCOLB PDF image created entirely on a computer by means of graphical software.

Specific Findings

Examination of the one-page (page 4/11 ) LFCOLB PDF image file reveals that the certificate page comprises a single flattened bitmap raster image embedded within the PDF file. Thus, unlike the (page 2/8) LFCOLB PDF image, the (page 4/11) LFCOLB PDF image is not constructed from nine different layers.
The WH LFCOLB and the (page 2/8) LFCOLB PDF images are each composite images comprised of multiple layers when viewed in Adobe Illustrator. Each LFCOLB PDF image contains nine layers. The (page 2/8) LFCOLB PDF image also contains one additional layer for the Document 15-1 case label.
The nine layers include one background layer, six text layers and two White-splotch layers. The White splotch layers contain clumps of near-White pixels of the same size and resolution as the smaller text-layer pixels. All eight of the text and White splotch layers have transparent backgrounds.
The (page 2/8) LFCOLB PDF image is a mixture of two different pixel sizes. All the pixels are square. The edge lengths of the smaller pixels are half that of the larger pixels. The larger pixels are 150 PPI resolution and all belong to the Green basket-weave background layer. The smaller pixels are 300 PPI resolution and belong to the six text layers and the two White-splotch layers. The larger background pixels are Grayscale and variable in color and the smaller text pixels are Binary and are monochrome and dark Green-Black in color.
The flattened (page 4/11) LFCOLB PDF image is comprised entirely of square pixels of 150 PPI resolution. All the pixels of the (page 4/11) LFCOLB image are Grayscale and variable in color. All of the printed, typed, stamped and handwritten characters are a variable Green-Black color. The entire (page 4/11) image has a noticeably Greener tint than the (page 2/8) LFCOLB image.
Because the The (page 4/11) LFCOLB PDF image is flattened then, unlike the (page 2/8) image, the various elements of the (page 4/11) LFCOLB image cannot be individually selected in Adobe Illustrator and moved around on the page. Consequently, unlike the (page 2/8) image where they were movable, the state Registrar’s certifying elements (date and signature stamps) are fixed on the page of the (page 4/11) image.
The (page 4/11) LFCOLB one-page PDF image file also contains 21 added (and hidden) geometric objects, each occupying a separate layer in Adobe Illustrator. These are added object layers which are placed above the flattened LFCOLB image layer in the layer stack. These added layers include 12 line segments, two broad-line strikeouts and seven Black redaction rectangles. The line segments are various shades of Green-Black. The pixels of each of the 21 added geometric images are congruent with a rectangular grid of 150 LPI.
Both the (page 2/8) LFCOLB PDF image and the (page 4/11) LFCOLB PDF image exhibit a geometric precision that is not expected from a optically scanned image of a paper document.
Each and every pixel which comprise the (page 2/8) LFCOLB PDF image is congruent with a rectangular grid of 300 LPI. Each of the edges of each square pixel is congruent with the grid lines. The pixels which comprise the image of the background layer are 150 PPI resolution. The printed form lines and characters, and the stamped, typed, and handwritten characters which are embedded within the background layer are all Grayscale. Each “background” pixel is of course also congruent with a rectangular grid of 150 LPI. The square pixels which comprise the six text layers are half the size of the background pixels. These include the pixels of the printed, typed, stamped and handwritten characters. The text pixels are all binary and monochrome. The edges of these pixels are congruent with a grid of 300 LPI. The near-White pixels of the two White splotch layers are also monochrome and are congruent with a rectangular grid of 300 LPI.
The (page 4,11) LFCOLB PDF image is entirely comprised of uniform-size pixels. Each and every pixel is congruent with a rectangular grid of 150 LPI. Each and every edge of each of the square pixels which comprise the (page 4,11) LFCOLB image are congruent with the lines of a rectangular grid of 150 LPI.
Importantly the (page 4/11) LFCOLB PDF image does not appear to be a scanned image of a paper document containing a mixture of two different pixel sizes which would have been the case had a printed copy of the (page 2/8) PDF image been scanned to produce the (page 4/11) LFCOLB PDF image. The (page 2/8) LFCOLB PDF image contains a mixture of large pixels of 150 PPI resolution and smaller pixels of 300 PPI resolution.
The (page 4/11) LFCOLB PDF image has uniform pixels of 150 PPI resolution.
Examination of the (page 4/11) LFCOLB PDF image at a Zoom magnification of 6400% reveals no evidence of the sharp contrast differences between the Grayscale and binary text that would be present if the (page 2/8) LFCOLB PDF image had been printed and scanned to produce the (page 4/11) LFCOLB PDF image.
Because the actual work flow used to create the (page 4/11) LFCOLB PDF image is not known, it is not possible to state with absolute certainty that a laser copy of the (page 2/8) LFCOLB was not scanned to produce the flattened (page 4/11) LFCOLB PDF image layer.
This uncertainty arises because, in this case, the work flow would necessarily utilize both a laser printer and an optical scanner. Consequently, it is not possible to know with certainty the forensic details of the laser printed copy or, much less, the final pixel image produced by the scanner.

The actual vendor and model of laser printer that would have been used to print the (page 2/8) LFCOLB PDF image is unknown. The potential range of variability in the printed image is vast because of the great number of different laser printers and different imaging technologies in current use. The fact that the digital image is in color also adds to the uncertainty and potential variability of the laser printed image.
Laser printed copies made with the older equipment and technologies consist of relatively coarse halftone dot patterns on white paper with resolutions of up to 1200 DPI. The current state-of-the-art commercially available high resolution laser printers produce images of 2400 DPI X 2400 DPI resolution. The image models associated with these state-of-the-art printers are very complex. Recently, a team of researchers in Singapore claim to have achieved the highest possible resolution of 100,000 DPI for color laser printing.
http://www.theverge.com/2012/8/13/3239154/highest-possible-resolution-color-laser-printing-sin gapore
This resolution is thought to be the theoretical limit for laser printers.
A trial laser print of the (page 2/8) LFCOLB PDF image was produced using a B&W Hewlett Packard Laserjet 1320 laser printer. The selected resolution was 600 DPI. This image was then examined at 8X magnification. The image consisted of a distinct Black-dot halftone pattern. The line screen orientation was ±45°. The 300 DPI monochrome text was uniformly Black with no detectable dot pattern within the interior of each character. The 150 PPI Grayscale text exhibited a distinct Black-on-white dot pattern within the outlines of each character. All the text characters were surrounded by white halos. Had this been a color print (made on a color laser printer of the same vintage), then three additional half-tone dot pattern images would have been laid down in Cyan, Magenta and Yellow along with the Black.
Additional uncertainty arises when the printed color copy is optically scanned to produce the (page 4/11) LFCOLB PDF image. The scanner basically reads the halftone dot pattern of the printed copy and converts the halftone image to a digital display image in pixels. The METADATA extracted from Document 35-1.pdf indicates that a Fugitsu ScanSnap S1500 color flatbed scanner was used to create this four-page PDF document. This particular scanner (together with it’s management software and a supplied third-party OCR program) can directly produce a PDF text file which is searchable . This Fugitsu scanner is also shipped with Adobe Acrobat X as bundled software.
The color values of the (page 4/11) LFCOLB PDF image were measured in Adobe Illustrator at the single pixel level. The results indicate that the RGB color values vary throughout each pixel of the (page 4/11) LFCOLB PDF image. The color values are nowhere constant except within the bright Blue text characters of the added second case label. These findings apply to both the text, form, handwriting, and background pixels. The color values of this second case label are R = 0, Red = 0, Blue = 255, indicting the pure Blue color of the RGB standard. The altered first case label is comprised entirely of multicolor pixels.

Except for two of the text layers, all other text (and White splotch) layers contained within the WH LFCOLB PDF image and the (page 2/8) LFCOLB PDF all have different monochrome colors. All of the pixels comprising the Green basket-weave background are variable in color. Only a few scattered characters belonging to the printed form, and typed text which are embedded within the Green background layer have variable color values.
Additionally, portions of the signatures and the last number “1″ of the Bates-stamped BC number 151 61-10641 are also embedded within the background and are thus variable in color.
When viewed in Adobe Illustrator, each text layer contains monochrome text on a transparent background. Nevertheless, within each text layer the same monochrome color is assigned to both the text and the transparent background. With the exception of only two of the six text layers, a different monochrome color is assigned to all the other text layers. The near-White colors of the two White-splotch layers are also different. However, the optical scanner would have to default to the variable background color when scanning the transparent regions of the text (and White-splotch) layers or else these layers would appear as opaque redaction layers in the final pixel image.
Additionally, the scanner would have to assign a continuously variable and different color to each pixel of the (page 4/11) LFCOLB PDF image from scanning a paper color printout of the (page 2/8) LFCOLB PDF image where the color which is assigned to each text (or White-splotch) layer is a monochrome color.
Another significant finding which is common to the WH LFCOLB, the (page 2/8) LFCOLB and the (page 4/11) LFCOLB images is the complete absence of color fringes adjacent to regions of sudden change in contrast. Color fringes are found next to high-contrast edges of objects within the image. These fringes are caused by lateral chromatic aberration arising within the optical elements of the scanner.
In text documents, the color fringes are seen along the edges of the text characters. The color fringes are usually either red and blue or magenta and green in color. The different colors appear on opposite sides of a region of sudden step change in contrast like the vertical strike of a character or the inner and outer edges of a loop in a character. The color fringes can be located either at the top and bottom of each character or on the left and right sides of each character depending on the scanning direction in relation to the scanned page.
The complete absence of color fringes in a scanned image is unusual but not impossible. Some very high-quality scanners can produce scanned images with little or no chromatic aberration. This is usually the case with archival copy quality scanners. Even in some of the best of scanners, the color fringes are removed by the scanner management software during the scan.
Mid-priced scanners which are more likely to be used for production scanning in an office environment typically yield color fringes on scanned copies. For example, the Associated Press scanned one of the handout copies of the Obama LFCOLB which the White House released to the press corp on Apr. 27, 2011. This scanned AP LFCOLB image has color fringes.

The chromatic aberration in digital images can also be removed by post-processing the image with special software on a computer. The Adobe Camera Raw (ACR) and Adobe Lightroom (LR) programs have color fringe removal ability. The ACR software is used with Adobe Photoshop (PS). However, this software removal technique is most often applied to digital photographic images rather than with scanned document images.
One would not expect this post-processing to have been applied to the (page 4/11) LFCOLB image if it had been scanned in a mid-quality scanner in a typical office environment.
Collectively the findings reported herein indicate that the (page 4/11) LFCOLB PDF image was not created by scanning a laser printed copy of the (page 2/8) LFCOLB PDF image.
However, this conclusion presents a dilemma because the METADATA extracted from court document 35-1.pdf, which contains the (page 4/11) LFCOLB PDF image, reveals that Document 35-1.pdf was created by a Fugitsu ScanSnap S1500 flatbed document scanner. This is a mid-grade color scanner which might be found in a typical office environment.
Consequently, the (page 4/11) LFCOLB PDF image was either created by scanning some other copy of the Obama LFCOLB (other than either the WH LFCOLB or the (page 2/8) LFCOLB) or it was created entirely on a computer using a graphics software program. If the latter, then the file METADATA is not consistent with the document 35-1.pdf work flow because the document indicated that it was created by the Fugitsu scanner. If the former, then the source of and evidentiary value of the paper original that was scanned is not known.
It follows that this situation is not unlike the typical circumstantial legal case where the entire body of circumstantial evidence must be weighed against the uncertainty in the circumstances of the case.

Exhibit 1
In addition to the single flattened LFCOLB image layer, the (page 4/11) LFCOLB PDF image file has 21 added geometric objects in separate layers. When the interconnecting objects and the second case label are also counted, there are a total of 40 layers added above the flattened LFCOLB image layer.
One important finding is that all of the 21 added geometric objects can be associated with a single transparent overlay page. These 21 objects are associated because their visibility is controlled by the same Compound Path object and they are grouped by the same Clipping Mask object into a Clip Group. The following steps will accomplish the task in Adobe Illustrator CS6.
1. Open the Document 35-1.pdf in Adobe Illustrator.
2. Select page 4.
3. Open the layers panel.
4. Select the topmost Clipping Mask in the layers stack.
5. Turn off the Compound Path object directly beneath the Clipping Mask object.
The 21 geometric objects are then visible and can be selected as a group and moved around the page of the flattened LFCOLB image layer. It is not necessary to first group these objects because they are effectively grouped by the Clipping Mask object. The redaction page (and the 21 geometric objects) can be moved by selecting any one of the 21 geometric objects and dragging the selected object to a new location.
The 21 hidden objects are made to be visible (or hidden) by switching off (or on) the one Compound Path object. However the 14 line segments can be made visible only if the Clipping Mask object is first selected. If the Compound Path object is turned off when the Clipping Mask is not selected, then only the seven redaction rectangles are made to be visible.
This off/on switching action is preserved even when the overlay page is displaced to a position entirely outside the rectangular boundary of the Compound Path. The rectangular boundary of the Compound Path is congruent with the rectangular boundary of the 8.5 in X 11. in LFCOLB page. Thus the entire overlay page can still be made invisible after it is slid entirely off the flattened LFCOLB page by switching the Compound Path object back on.
Exhibit 1 shows three full-page frames of the (page 4/11) LFCOLB PDF image. The top left picture shows the (page 4/11) image as it would appear in Adobe Reader. The top right figure shows the redacted image with the 21 hidden geometric objects switched on. The hidden image is revealed when the one Compound Path object in the layers list is switched off in Adobe Illustrator. The outline of the redaction overlay page is also shown in this picture.

The redaction overlay page is smaller in size than the LFCOLB page; it is smaller in both width and height. The width of the overlay page is approximately the same as the width of the Green basket-weave security paper background of the LFCOLB image plus the width of the lefthand margin of the LFCOLB page. The left edge of the overlay page is congruent with the left edge of the LFCOLB page. The top edge of the overlay page coincides with the top edge of the certificate portion of the LFCOLB page. The bottom edge of the overlay page is close to (and parallel to) the bottom edge of the Green basket-weave background of the LFCOLB image. The right edge of the overlay page is coincident with the right-hand border of the basket-weave background.
The bottommost image in Exhibit 1 shows the redaction overlay page slid off the right side of the flattened LFCOLB page. When the Clipping Mask is selected, all of the 21 added geometric objects are fixed within the redaction overlay page and move as a group relative to the LFCOLB page. However when this selection is turned off, any one of the seven redaction rectangle objects can be individually selected and moved over both the page of the flattened LFCOLB image and the redaction page. (The Clipping Mask can be most easily deselected by turning off its layer and then turning it back on.) However, when one of the redaction rectangles is selected and moved, then only the outline of the redaction rectangle is relocated and the (portion of) monochrome interior of the redaction rectangle (which is outside of the original redaction rectangle) instantly turns to transparent. If the displaced position overlaps the original position then the overlap region is Black and the remaining portion of the displaced rectangle is transparent. The remaining portion of the un-displaced rectangle is also instantly transparent. Only the overlap region is Black. When the displaced rectangle is completely outside of the original position of the same rectangle both the displaced rectangle and the original position of the rectangle are entirely transparent. The displaced position is located by its Blue rectangular boundary. The initial position of the rectangle is then transparent and without a boundary.
Curiously, if one redaction rectangle is moved to a position where it doesn’t overlap the initial position and then the Clipping Mask is re-selected, then the redaction rectangle’s Blue boundary instantly displaces back to its original position and the interior of the redaction rectangle is thereafter entirely transparent. The group of 21 objects can then be displaced and the Blue boundary of the (now transparent) redaction rectangle (which had been previously selected and moved) moves with the other objects. However, if one redaction rectangle is moved to an overlapping position and then the Clipping Mask is selected, then the Blue boundary (as before) instantly displaces back to its initial position. However the overlap region is then Black and the remainder of the original rectangle is now transparent. The Blue rectangular boundary is back to its original position. Then if the group of 21 is selected and displaced, the partially-opaque/partially- transparent rectangle moves together with the group of 21.
If a large redaction rectangle is selected and moved to a position which overlaps a smaller redaction rectangle then the overlap region remains Black. If a large rectangle is displaced to a position which overlap its initial position and also another rectangle, then the interior of the displaced rectangle contains two Black overlap regions. The remaining portion of the interior of the large rectangle switches to transparent. The portion within the original position of the large rectangle which has been uncovered is also transparent.

The 21 hidden geometric objects exhibit a geometric precision that is not expected from a optically scanned image of a paper document. The rectangular boundaries of these 21 objects are congruent with a rectangular grid of 150 LPI. Each side of each and every rectangular object is congruent with the lines of a grid of 150 LPI. Consequently, each object contains an integer number of pixels and each side of each rectangle consists of an integer number of side-lengths of a pixel of 1/150 in X 1/150 in square dimensions.
An additional finding is that each of the seven Black redaction rectangles was constructed from the same single Black pixel of 1/72 in X 1/72 in square dimensions. Each rectangle was constructed by scaling the sides of this single Black pixel. The X and Y scale factors for each redaction rectangle were read off the links data in Adobe Illustrator. The scale factors are consistent with the integer number of pixels for a resolution of 150 PPI X 150 PPI for each rectangle. Consequently, the scale factors are not integers when the scale factors are not expressed as percentages. The 1/72 in X 1/72 in pixel is the original standard size of the screen pixels of the Apple Macintosh Computer. Each redaction rectangle is an embedded bitmap object in Adobe Illustrator and the objects’ file links to the external image file of the single Black Apple Standard pixel are all missing.
Clearly, the 40 added layers including the 21 geometric objects were constructed by someone with an expert knowledge of Adobe Illustrator and with a particular hidden purpose in mind.

Case 3:12-cy-0020.04-FrALLRA Dccument 35-1 Filed OW05112 rage 4 of 4
:AN!? .::1-,.-nv-64.1.1(1-F1 I VP-L1….. L,..111 -IC111 1::-.
Pel—Tore 011eFIFICILTI OF LIVE MTH 1311241_1,11111 1,1,TILL
61 11:1611.
SI. VC PLY.ITIR ODOM, 1:
-p- 34 • 1. ‘1,4 aroo i Is ea….-1,a. • . Ma De IC . IW
,.. 13 e marm . , .aa • 1..e …ma P.I.•
mg* ,..,..a. -na.0 Taa•13 .6 mr_D: … 4. 1 541 7.71 7.,
Faual.f c r..i.
acir=-Jr1F-.. —L761—
l41=3. npippraPp 4, Frr7,47:1:17•71. !sane W. 1 Llat 0 ….
4.a…… a……..–.- – wad
Ekrtioull. halm imatalls. m
Irni_l.
4- ‘…1’ – 1a7 I oralaGWa.14.7.3170—
MP gagaAlia.-‘1. miugy I ……e-reig
1 1-…–…–,..
.771- 77121 –
=pt..
.793 MAY.
54,71.1P. LW IMITAPI PIRICIMILMI
ri- C..’ .*. o• cc* ..c. c • oc..*, Pc or …… -4ma. Pm. Prm- rra-• . IA Lc Li Ie.
=-•- •-… .4.0
• . Ire, er c•••••• ….,…a, ….7
-0’1 Mi la marmm F.E.•
,… • rt . .’ 1:r..i. .
1 6
.. .- 4,………e…– zsza,..1 iL-7-41
r
=.1.-}”iT:a 41;71E.Gt,L44.12,4,,,,i, paa9. r i 11
rs,….-1,zr–

Casa 3:12-cv-120260-FIPALLRA Dccdmerd 35-1 Filed 0EU06112 Page 4 of 4
:anfi :-.1ii.-ny-LitIL1:L1-hl I LT111-1C111 Lt..%,4?”
011111.101k OF 1.1111, =TX 156.101.111.11
gl /101G41

1.1•Ift
ir•
• r
Ekmalnlzi
MAP V-4441.1a,-14 114/1..7 7C-Ni•PL11…,*

PFR 2011 C.
itrciiankri AFR 21, 2011 r=ro-oll
a-4m riccinT41

Exhibit 2
The Green basket weave safety paper background of the (page 4/11) LFCOLB PDF image appears to have been patched in both lower corners of the page. It is not known whether this was a paper patch applied to a real paper document or whether these patches were simulated in a computer graphics program.
Exhibit 2 shows a comparison between the lower corners of the WH LFCOLB PDF image and the (page 4/11) LFCOLB PDF image. The top two frames in Figure 2 compare the bottom left corners of the two images and the bottom two frames compare the bottom right corners. In each pair of frames the WH LFCOLB is on the left and the (page 4/11) LFCOLB is on the right.
There is a noticeable difference in color between the patched regions and the remaining background in the (page 4/11) LFCOLB. Additionally there is noticeable misalignment of the bars of the basket-weave pattern along the edges of the patches. These misaligned bars have the appearance of wrinkles in the basket-weave pattern near each corner. None of these artifacts are detected in the WH LFCOLB image.
Most importantly 3 of the 21 hidden geometric objects are associated with these patched regions. These three objects are two line segments and one Black redaction rectangle. These objects are also shown in Exhibit 2. The two vertical line segments are placed so as to locate the leftmost edge of the page and the left edge of the Green background of the bottom-left patch. The vertical position of the top end point of the vertical line segment which coincides with the left edge of the Green background is aligned with the top edge of the bottom-left patch. The bottom end point of this same line segment coincides with a small offset step in the vertical edge of the Green background of the patch near the bottom-left corner. One of the Black redaction rectangles is within the interior of the bottom-right patch.
Clearly the two hidden vertical line segments were placed to mark the position of the bottom-left patch.
Possibly the Black redaction rectangle was used to locate whatever was later covered up by the bottom-right patch.
Because the basket weave safety paper used to copy certificates by the HDOH is 24 lb heavy-weight stock, the presence of real physical wrinkles in the safety paper of a certified copy of an original LFCOLB is unlikely. Moreover, the observed deformation of the basket-weave pattern is of a shearing type which would be difficult if not impossible to achieve in an intact sheet of heavy-weight safety paper unless it is torn.
The safety paper routinely used by the Hawaii HDOH is reported to be “Design Secure” Safety Paper purchased from “Simpson Specialty Papers”, a company that was founded in 1871. One of the security features of this particular paper is that it has invisible fluorescent security fibers randomly dispersed within the paper. The fibers are invisible under ordinary White light. However, when the certificate is illuminated under Black light (Ultra Violet light) the colored

fibers fluoresce and become visible. In this light, the colored fluorescent fibers glow brightly and appear as short random Green lines in the paper. Consequently, the authenticity of the safety paper used to produce a genuine HDOH certified copy can easily be verified by means of a hand-held Black Light source.

bilth-certificate-lanikfonr…pdf M (R(B/PFeview X 35-Lixr @ 330% (RGE/Preview)

Exhibit 3
The two shortest line segments of the total of 12 hidden within the one-page PDF file of the (page 4/11) LFCOLB are both associated with a visible line within the flattened (page 4/11) LFOLB image. This single vertical line spans the height of the certificate portion of the LFCOLB page next to the gutter region. The length of the longer of the two short line segments is one pixel greater than the thickness of the vertical line. The shorter of the line segments spans half the width of the line. The longer of these two short line segments is located at a vertical height from the bottom edge of the certificate portion approximately 3/4 up the length of the vertical line. The shorter of these two line segments is located at a height near the midpoint of the vertical line.
Although the two hidden line segments are very short line segments, each appears as a dark-Green rectangle in the layers list. The solid color assigned to each short line segment nearly matches the color of the vertical line at the location where they are placed. These shortest line segments are actually tiny rectangles. Each green rectangle becomes visible when each line segment is selected and the Compound Path object is turned off. It is necessary to also turn off the largest Black redaction rectangle in order to clearly see the boundaries of the shorter line segment.
Exhibit 3 shows these two shortest line segments and the adjacent portions of the vertical line upon which they are placed. These findings suggest that these two short line segments were possibly used as an aid in the placing of the vertical line within the certificate portion of the LFCOLB image. Or possibly these two short lines could have been used to locate the vertical height of other features within the certificate. Alternatively they could have also have been used to align, to position or to size the flattened LFCOLB image with respect to the redaction overlay page.
Actually, each hidden geometric object and the redaction overlay page could have been used to position, align, size, or assemble the various elements of the LFCOLB image onto its 8.5 in X 11. in blank page. The left edge of the redaction overlay page would be brought into coincidence with the left hand edge of the LFCOLB page. The top edge of the overly page would be made to coincide with the top edge of the certificate portion of the LFCOLB image layer. The bottom edge of the overlay page would then be close to and parallel with the bottom border of the Green background of the LFCOLB image. In effect, the redaction overlay page could have provided a template for re-sizing the LFCOLB flattened image from the scanner.
These findings prove conclusively that the (page 4/11) LFCOLB PDF image file was not created by simply scanning a paper document. Instead 14 hidden line segments and 7 Black redaction rectangle objects and their associated interconnection objects were added to the PDF file over a flattened LFCOLB image layer. A total of 40 layers were added.
If instead, the (page 4/11) LFCOLB PDF image was created by scanning some other paper copy of the Obama LFCOLB (other than a printout of the (page 2/8) LFCOLB image), then the two case labels added at the top edge of the (page 4/11) LFCOLB PDF image would be deceptive.

Exhibit 4
The two topmost layers of the WH LFCOLB PDF image file contain only scattered clumps of near-White pixels. However, these scattered “White splotches” are entirely missing from the (page 4/11) LFCOLB PDF image.
Exhibit 4 shows a comparison of these two rectangular White-Splotch regions between the two LFCOLB images.
These two frames prove that the White splotches that appear as separate layers of the (page 2/8) LFCOLB PDF image are not present in the (page 4/11) LFCOLB PDF image.

Exhibit 5
There are only four typed letters which have the same pixel resolution between the WH LFCOLB PDF image and the (page 4/11) LFCOLB PDF image. These are the only typed characters that appear on the Green basket-weave background layer of the WH LFCOLB. Hence these four characters are the most likely to be identical between the two LFCOLB images.
These four characters are purported to be uppercase typed letters. The four letters are the “R” in BARACK, the “K” in Kenya, the “S” in Stanley and the “K” in Kansas. The resolution of these four letters is 150 PPI in both LFCOLB images. The “R” and “S” letters were compared between the two LFCOLB images. The letter “K” in Kenya was not compared because the vertical strike of the character overlapped a vertical form line. Similarly the letter “K” in Kansas was not compared because the top half of the letter was missing in both LFCOLB images.
Exhibit 5 are magnified images of the “R” and “S” letters from both LFCOLB images. The top two frames in Exhibit 5 compare the letter “R” and the bottom two frames compare the letter “S”. The frames on the left are from the WH LFCOLB PDF image and the frames on the right are from the (page 4/11) LFCOLB PDF image.
Based on these frames both characters are significantly different at the same pixel resolution of 150 PPI. One of the most noticeable differences is the disparity of colors of the pixels in each letter. Another difference is that the “White Halos” that surround each character of the WH LFCOLB and the (page 2/8) LFCOLB are missing in the (page 4/11) LFCOLB.

birth-certiticate-long-form.pdf 4800% (Ra/Preview) x
IN MINIM
N EM MINIMS =NNW NOMINEE
N OMINEE NOMINEE NOMINEE
E MMEN
MINIM MIMI MINIM MINN MIMI MIMI
N ENN NENN IMEN MINIM NENN NEM
IIMEN MINIMS INEMEMENEMENE INEMENEMENINE INNENNENNENNEN IIIIMINEMENINE
IUUU NNW
ION MIMI
INN NENN
IIMEN MINIM
INNEN MINIUM OMEN MENOMONIE INNEN =NNW MIMI
N NW MINIM =NNW MEMNON MENNEN
E NEMIES IMMO IMO NENN NEN IMMO IMO INN NMI MIN MIN
N EN NEN MIN MIN
N EN NEN MINN MIMI IMENNINEMENE MENOMONIE.
N INNENNENNENNN MINIM MINIM
N INN NENN
MIN MIN
N EN NENN
MINIMS MINIM MINIM MENNEN MINION ENEMIES MIMI NOMINEE MINIM
NENN NE NMI IN


11 11
I
II IU
Eh IN MI

birth -certificate-long-fomipif @ 4800% (RGB/Pieriew)

Exhibit 6
Exhibit 6 shows the same uppercase “O” character from each of three aforementioned LFCOLB PDF images. The landscape page has been rotated 90 degrees counterclockwise in Exhibit 6. The left-hand frame is from the (page 2/8) LFCOLB PDF image, the middle frame is from the (page 4/11) LFCOLB PDF image and the right-hand frame is from the AP LFCOLB PDF image. The AP LFCOLB image is a scanned copy of the WH press corp handout copy of the Obama LFCOLB.
A significant finding is that the shape of each text character of the (page 4/11) LFCOLB PDF image is distorted. The distortion is the “pincushion” type where the horizontal and vertical lines curve toward the center of each text character. This effect is revealed by a slight inward curvature of the grid lines near and passing through each character. The effect is most easily seen by examining the shape of the letter “O” wherever it appears. The rectangular grid lines are straight throughout the background regions and in most of the form lines. (See Exhibit 6).
Additionally, the AP LFCOLB PDF image also exhibits the same pincushion distortion of each text character. (see Exhibit 6).
This pincushion distortion of the text characters is not present in the WH LFCOLB PDF image or the (page 2/8) LFCOLB PDF image. The rectangular grid lines are everywhere straight in these earlier LFCOLB images. (See Exhibit 6)
This particular distortion is unusual and is not a byproduct of optical scanning of a paper document. Rather it is more likely a product of some unknown software post processing of the digital image.
These findings prove that the (page 4/11) LFCOLB image and the AP LFCOLB image both exhibit pincushion distortion of each text character but the WH LFCOLB and the (page 2/8) LFCOLB do not.
This finding also supports the conclusion that the (page 4/11) LFCOLB PDF image was not created by simply scanning a laser-printed copy of the (page 2/8) LFCOLB PDF image.

Exhibit 7
Another finding is that the (page 2/8) LFCOLB PDF image includes a faint image of the purported Hawaii raised seal impression but this seal impression is totally missing in the (page 4/11) LFCOLB PDF image.
Exhibit 7 compares the region of the seal impression between the two LFCOLB PDF images. The landscape page has been rotated 90 degrees counterclockwise in Exhibit 7. The frame on the left is from the (page 2/8) LFCOLB PDF image and the frame on the right is from the (page 4/11) LFCOLB PDF image. The outer rim of the seal impression is seen as an irregular circle of white dots in the left-hand frame. The second White-splotch layer overlaps a portion of the circle as shown by the Blue rectangle. The circle of dots is entirely missing in the right-hand frame.
This finding also supports the conclusion that the (page 4/11) LFCOLB PDF image was not created by scanning the (page 2/8) LFCOLB PDF image.

Henry Blake, PhD
02/19/2013

10 thoughts on “MS – Orly v Democrat – Henry Blake Affidavit – Part 1

  1. What the heck does that all mean? Anything?

    Well, the first document the Democratic Party submitted was exactly the same as the Whitehouse PDF. The second document is a scanned version of the first document and was imported into the PDF using paper capture plugin. So far I have seen nothing that suggests anything nefarious.

    The blocks and lines are interesting but they are invisible and I believe they may have been generated by the scanning solution. OCR was also turned on, which made the document searchable.

  2. WTF are the parts where he starts writing in tongues? I swear I thought he mentioned Ezekiel but it was actually Ekmalnlzi (wasn’t he a Sumerian god along with Zuul and Gozer) and another time I thought he was saying eeny, meeny, miny, moe over and over again.

    Can you translate that?

  3. Well, the documents itself show little more than a few minor scanning artifacts. I guess anyone with Illustrator considers himself nowadays an ‘expert’?

  4. All I get from this overly long beyond-pixel-peeping drivel is:

    “Your honour, there are two photographs of the same building, but the JPG’s are not bit-wise identical and there is some additional shadow in the second photo, THEREFORE FORGERY.”

Comments are closed.