(Archived) OCR of pdfs that include handwritten text

[mclovin 1]

I've looked at a lot of posts on these forums and learned quite a bit about Evernote and different kinds of pdfs, etc. But I can't quite figure out the answer to my particular situation and question.

If I write a sheet of handwritten notes, use an MFD or a scanner to scan the notes to a pdf file, then put that in Evernote (premium account), Evernote will try to OCR the pdf but it will not recognize any words and isn't searchable. Now, if I scan the document into a tiff file, Evernote will OCR it and recognize words and it therefore becomes searchable for me.

My questions are:

- what will Evernote recognize if I scan a document of typewritten words into a pdf file and write a few handwritten notes at the top of the page? Will it recognize the type, or the handwriting, or possibly both?

- How about if I generate the document from within an app like Word so its the kind of pdf that has a text layer in addition to the image layer?

- Is there any difference between how this all works on Mac or PC?

My goal is to take a page of sloppily handwritten notes which the best OCR system in the world could not possibly handle, but legibly write a few key words (like tags) at the top of the page. I want to scan that into some file format, pop it in Evernote and have it recognize the legibly words I write at the top so I can search by those.

Thanks for any help anyone can provide!

Jim

[jbenson2 2,147]

My goal is to take a page of sloppily handwritten notes which the best OCR system in the world could not possibly handle, but legibly write a few key words (like tags) at the top of the page. I want to scan that into some file format, pop it in Evernote and have it recognize the legibly words I write at the top so I can search by those.

Rather than rely on my attempt to write legibly, what I do is put the handwritten note to Evernote, then add some typewritten comments and tags above the PDF file in the same note. This way, I am sure that there won't be any question by Evernote on what the words are supposed to be.

[BurgersNFries 2,407]

viewtopic.php?f=30&t=15182&p=59616&hilit=handwriting#p59616

[mclovin 1]

My goal is to take a page of sloppily handwritten notes which the best OCR system in the world could not possibly handle, but legibly write a few key words (like tags) at the top of the page. I want to scan that into some file format, pop it in Evernote and have it recognize the legibly words I write at the top so I can search by those.

Rather than rely on my attempt to write legibly, what I do is put the handwritten note to Evernote, then add some typewritten comments and tags above the PDF file in the same note. This way, I am sure that there won't be any question by Evernote on what the words are supposed to be.

Thanks, that's a good idea that. I'm actually testing out a specific use case for a VIP in my company who would likely not having to go the extra step of adding type written comments and tags to the pdf. Silly, I know, because I'd be fine with it myself.

Appreciate the reply.

[mclovin 1]

http://forum.evernote.com/phpbb/viewtopic.php?f=30&t=15182&p=59616&hilit=handwriting#p59616

Yeah, this is one of the threads I saw. I'm really trying to get a straight up yes/no answer though which I don't think I've really seen in any of the threads. I see a lot of 'it's best to' and 'try to scan as an image' etc (which is good because it does help), and I've learned a lot about the differences about pdfs and how Evernote handles them in most cases but I can't seem to get a yes or no.

For example, I have a page from a company's annual report and I've written 'Important' at the top of the page. I scan that as a pdf and put it into Evernote - will it recognize ANY text in that file?

I'm researching Evernote for wide spread usage in our company so it's important that I get a concrete yes or no answer on this one, but also really understand how it's working.

Thanks!

[engberg 89]

Handwritten text in PDFs will only be recognized if it is very clear and printed. I.e. the OCR engine we've purchased to process PDF documents doesn't do much with handwritten text.

[mclovin 1]

Handwritten text in PDFs will only be recognized if it is very clear and printed. I.e. the OCR engine we've purchased to process PDF documents doesn't do much with handwritten text.

Gotcha. I can work with that answer. Thanks again, Dave.

Jim