(Archived) pdf and OCR scan

[manfreed 0]

hi!

I understood that when i paste a pdf in a note on evernote that this document will be scanned on the evernote server (for premium users immidiantly) and I will bei ably to search this text on the pdf (as it works with pictures). is this not right?

thanks for helping and for this great programm (special for iphone user)

manfred

[engberg 89]

You can currently search for the "normal" text within the PDF documents in Evernote. I.e. if you can select the text from your favorite PDF viewer and copy-paste it, then it's normal text within the PDF that we'll process.

If your PDF contains images, and those images have text printed on them, we will not yet search for that text. We would like to make this happen in the future, but it's a bit complicated since we've seen many PDFs that have hundreds of different images in different low-level data formats. This presents a bit of a processing and UI challenge for us to index and then show matches later.

Thanks

[manfreed 0]

oh....

but it works if I save the pdf as a jpeg file.

thanks

Best greetings

manfred schindler

[aashish108 11]

You can currently search for the "normal" text within the PDF documents in Evernote. I.e. if you can select the text from your favorite PDF viewer and copy-paste it, then it's normal text within the PDF that we'll process.

If your PDF contains images, and those images have text printed on them, we will not yet search for that text. We would like to make this happen in the future, but it's a bit complicated since we've seen many PDFs that have hundreds of different images in different low-level data formats. This presents a bit of a processing and UI challenge for us to index and then show matches later.

Thanks

Hi, did you manage to solve this challenge? The text on image in a pdf? What's the update?!

Did a quick forum search, and there's this road block and that its too much of a strain on the evernote servers?

[engberg 89]

If you scan a document as a PDF and have a Premium subscription, we will process that scan by running OCR on those full-page bitmap "images".

If you have a text-based PDF with normal selectable/copyable/searchable text, and that PDF happens to contain a few embedded images or illustrations, we won't process those images.

I.e. the OCR feature is for scanned documents, not for trying to find a word or two in every little picture in a "normal" PDF document printed from Powerpoint, etc.

[aashish108 11]

Hi, Thanks for the info. What is the timescale for indexing image-based pdfs via OCR? I uploaded a lot of PDFs a couple of days ago, but it appears that indexing them will take a bit of time? I have a premium account.

Thanks!

[engberg 89]

If you have a Premium account, and you've uploaded scanned documents in PDF format, they will usually get processed within a few minutes.

You may want to check whether the documents you uploaded are scans that don't already contain text that you can select/copy/paste from Preview.

[aashish108 11]

Hi, thanks for the reply. The PDFs are definately image-based but I did not scan them, it was done some time ago. The text has not been OCR'ed as I tried searching for words that I see on the PDF but none of the PDF's show up in the search. It has been like this for several days now.

Any ideas? Also, I am using Evernote 4, the latest development version.

[engberg 89]

Can you open one of those PDFs in Preview and try to use the "Select All" menu option and then the "Copy" option?

Then try pasting into a text editor to see if you copied any text.

I'm curious whether your PDFs already contain some text that can be selected and copied. (Evernote doesn't process documents that already contain searchable text.)

[aashish108 11]

Nope! No text whatsoever. You can tell each page are images as each page looks distinctly scanned as the pages were quite old and it shows.