Small oddity found in DT3 OCR process of PDF files

DEVONthink
1

Hello everyone,

I have been using DTPO2 for a few years and now am evaluating moving to DT3, which would also come together with a few important changes in some workflows I have been using.

In my DT3 evaluation, I have come to appreciate the improved OCR engine, and in my tests I have found a little oddity in the PDF file before and after OCR processing.

I am using HP Easy Scan on High Sierra for the tests. I find that little utility very simple and fast, and with a higher quality and file compression ratio which are very nice.

As you can see from the “before the OCR“ file details, the size is A4, listed as 21.0 x 29.7 cm:

lux-screenshot0659
lux-screenshot0659524×724 43.5 KB

But after the OCR process, interestingly enough the page size has changed to 21.1 x 29.7 cm:

lux-screenshot0660
lux-screenshot0660502×726 42.6 KB

I have tried to also remove all the other options from the OCR menu, including file compression, page orientation and deskew, and the resulting page size is also 21.1 x 29.7 cm.

I wonder why this is the case. It’s not a huge issue, but I have noticed that the small change in page size makes PDF look slightly different in terms of size when opening the pre-OCR and post-OCR file.

Any clue what might cause this?

Thanks,

Luca

2

Development would have to assess this, but it could be nothing more than a simple rounding error.

3

I think this is caused by the ABBYY OCR when it is generating the new PDF. It should the same size as the original however I have seen occasional differences although none that have been consistently reproducible. I will raise it as an issue with ABBYY.

4

Thanks @aedwards let us know when you hear back from ABBYY about this.

Bye, Luca