Plugin: offline OCR (extract text from images, pdf, videos, etc)

Plugins
44

@Wimvan Yes, the third number in version means that it's a patch version. FYI, The second number in version means new feature contained.

1 Like
46

Great! I was going to write you about the text insertion, but I felt like you had already thought about it and you did. Thanks so much. I am looking forward to the next releases. Keep up the good work!

47

Thanks very much for creating this plugin! I'm converting over from Evernote and trying to learn the ropes and am a little confused about the correct way to use this plugin. Two questions please:

  • When I click on the OCR icon in an image or PDF and ask it to recognize the text, it works, and the next screen comes back with the text recognized. The only option at this point on the screen is "QUIT." What am I meant to do at this point? Does this mean the OCR text was found and inserted to my note in a way I cannot see? Or am I supposed to copy it from that screen and paste it somewhere in my markdown editor?

  • I think this is meant to be an offline editor, but it's not clear to me how it runs without manual effort (or if it does). Is there a way it will process all notes either as new notes are created or all notes with relevant attachments?

Thank you.

48

Hi & welcome!
My experience with that plugin is:

Yes. I can edit the recognised text in that little window and then manually copy it before I quit the screen. Then I paste it into wherever I want, in most cases a joplin note. There is no automatic process beyond the pure recognition of text.

Offline means that you don‘t need a connection to your network / the internet to be able to use the text recognition tool. Everything it needs is on your device.

I‘m not sure I get your question right, You can use the import function to transfer your notes from Evernote into Joplin. All attachments are included in this process. But there is no OCR done during import process.

2 Likes
49

Thanks for clarifying. You understood my question correctly and I appreciate the reply.

1 Like
50

thank you @ylc395 for this very helpful plugin!

It seems the auto-recognition is not working anymore, at least in my case (joplin 2.10.11, ocr 0.3.2).
Maybe this was deactivated when the plugin backend was updated?

Alternatively, automatically including the detected text in the image markdown link would be amazing!

Is the plugin still under active development?

51

I agree that losing OCR is a major barrier to leaving Evernote.

Current OCR plug in documentation says it only handles notes that have a special tag. It looks like ocr-joplin notes will attempt to OCR every note unless a tag is suggested.

Do you run it on a chron job or some other automatic execution?

I hope it is smart enough to avoid rescanning a note when the initial scan didn't detect anything OCR-able.

Thanks!

1 Like
52

The OCR v0.3.2 plugin seems to work fine for me. You do have to view the note in the Markdown view, but then in the top right corner of the image an icon overlay for OCR appears. Clicking it starts the OCR process.

So it is not automatic on all images like Evernote (unless perhaps there is a diferent plugin?), but it does work.

So it's not "automatic recognition" and neither does it insert the recognised text into the note. Am I missing something or is it broken in this respect?

53

So it's not "automatic recognition" and neither does it insert the recognised text into the note. Am I missing something or is it broken in this respect?

In the release notes for 2.11, the integration of plugin user data, such as OCR, is mentioned as an example application. Maybe the functionality is broken because of the restructuring of the plugin system? (What's new in Joplin 2.11 | Joplin)

In addition, the need for an OCR plugin is also expressed in the ideas for the google summer of code 2023. https://joplinapp.org/gsoc2023/ideas/

Maybe something is going to happen soon?