Jump to content
Join the #EverBetterChallenge and finish what you start in 2019. Read more... ×
TTTrouble

(Archived) Evernote OCR Not Working for Screencapped Images

Recommended Posts

Hello,

   I'm using evernote to study and taking screenshots of particular parts of PDF books that I use to study. Now I was hoping Evernote would automagically OCR the screencapped images, but I've checked on the notes I've took several days ago and they still haven't been put through the OCR recognition system. Am I right in believing 3-4 days is more than enough time for the OCR to work on a free user account? Are there some criteria that a screenshot must reach to be OCR'd? I guess I just can't figure out what could be causing the problem -_-.

 

      - TTTrouble

Share this post


Link to post

Hi - welcome to the forums. Do you happen to know what sort of file is created by your screen-shot software? Evernote does OCR JPG,  and (I think) PNG files - you'd see both of these screen shots as proper pictures in your notes.  If you're taking a screen grab you may have a lot of information on the screen - you'd need to make sure that what you can see is reasonably clear and legible.  If you're using a PDF reader to see your notes,  can you save the whole file,  or extract pages from the PDF file and save those?  If not,  you may need to enlarge the text and take grabs of a few paragraphs at a time, rather than whole pages or significant chunks of pages.

 

Having said all of which,  Premium users get first dibs on OCR work (sorry) and free users get processed in turn,  as a lower priority - 3-4 days could be the sort of time it will take if things are busy at home right now..

 

The way to test though is: do a screen grab of a few lines in the best quality you can manage and see what happens!

 

Edit: just to clarify - if you're not saving JPG (or -maybe- PNG) files,  the picture won't be processed.

Share this post


Link to post

Well I'm using the built in Evernote Desktop software where you can right click on the icon and click "clip screenshot" and you get some crosshairs to identify exactly what part of the screen you want to clip. I know for sure that the screenshot is saved as a PNG

 

4aixEwp.png

 

Looks like the text is big enough to me.

Even after a couple days, no real OCR recognition. There is an icon with a magnifying glass buried in the note information, so I think that means it has been scanned, the OCR just isn't able to recognize the type I guess.

Thanks so much for your help!

Share this post


Link to post

I'm not sure that OCR is quite ready for the multi-syllabic technical stuff that you're saving - I suggest you raise a support ticket (see below) to check whether anything is going actually wrong,  and see whether they can suggest any alternatives.  My view would be its a lot better to save technical stuff in PDF format and do your own OCR where you can build up your own specialised dictionary if necessary.

Share this post


Link to post

Plus, indexing text in images is different from indexing text. With images, EN makes a tree of possibilities. So an image with the word 'house' may show up when searching for the word 'horse'. If you want a more accurate indexing, rather than take a screen cap, you would want to either copy the text or have the PDF already indexed.

Share this post


Link to post
Guest
This topic is now closed to further replies.

×