Hermitian: Now maybe you Obots could explain how the PDF that produces the final images does not contain the final images but yet the final images just pop out when the file is opened in a bunch of different PDF editing programs? And please also explain why that same-sized final image sometimes pops out at 120 PPI x 120 PPI and some times it pops out at 200 PPI x 200 PPI whereas the two images that are detectable within the PDF are both 72 PPI x 72 PPI resolution?
Hermitian: W = 24.153 in. ; H = 28.764 in. (or W = 1739 P ; H = 2071 P). The smaller image is a certificate image with near-White background. The page size is W = 14.486 in.; H = 17.264 in. (or W = 1739 P ; H = 1243 P).
First of all, Hermitian has made a slight error here. the W of the second image is 1043 not 1739.
So let’s find out what is happening here. All lines up with the workflow I had described earlier.
First we look at the PDF from the Muscatine website and find the following PDF:
Which shows the following image information
8.7 × 10.36 inches 626.04 745.56 pixels
Internally the image is stored as
Obj 11 0
11 0 obj<</Subtype/Image/Length 294131/Filter/DCTDecode/Name/X/BitsPerComponent 8/ColorSpace/DeviceRGB/Width 1043/DecodeParms<</Columns 1043/Blend 1/VSamples[2 1 1 2]/HSamples[2 1 1 2]/Rows 1243/Colors 3/ColorTransform 1/QFactor 0.25>>/Height 1243/Type/XObject>>
Internally to PDF this object is /Im0 and is painted using Obj 10
‘q 626.0399933 0 0 745.5599976 0 0 cm /Im0 Do Q’
So what happens when you take a 1043×1243 image at 72 PPI and transform it to an image 626.0399933×745.5599976 inches? It becomes a 120 PPI image.. (1243/745.5599976*72 and 1043/626.0399933*72). This image is the one you are seeing when using a PDF viewer.
So while the PDF is at 72 DPI, the image is at a 120 PPI resolution. The same happens in the WH LFBC where the 72 DPI PDF contains a 150 PPI background and 300 PPI foregrounds.
This image can be extracted as 4db82608b486f-000.jpg which shows an Adobe header and the following dimentions
1043 × 1243 pixels 72 pixels/inch
However, there is also an embedded Private ‘image’ where Photoshop has stored its internal formatted PSD file information. This happens when a user checks the ‘allow editing’ when exporting to PDF. The object in question is a 5.5Mb data stream.
1739 × 2071 pixels 200 pixels/inch
I extracted this image using GIMP and it matches the AP1 image extracted by Hermitian.
Now do some simple math: 1739/200*120=1043 and 2071/200*120=1243
So now we can explain what happened.
The user imported the 200 PPI jpeg (2698×3234) obtained from the Associated Press or 13.49” by 16.17” and scaled/cropped the image to 1738×2071 (or 8.69×10.35” to make it closer to 8.5×11 format). This image was then used to correct for the bluish color imbalance. Once satisfied, it was used to create a PDF @120 PPI and the PSD file was stored internally.
PS there is also a small thumbnail <xapGImg:image></xapGImg:image>, 265×215 PPI
Understanding how Photoshop stores its ‘PSD’ in a private section of the PDF goes a long way to appreciating what happened.
Hermitian: Well NBC you are sharper than I thought. I too also noticed that the two images that I posted were the same size. But the difference between you and me on this particular mystery is that I noticed this on Jul 1, 2013. And I have been trying to figure out how the Hell this could happen for over a month. Which means of course, as always, I’m way ahead of you.
A month? Really? You being ‘ahead’ was because you allowed yourself to be side-tracked by this PDF. It took me little time and effort to move ahead and even overtake you.