[GH-ISSUE #424] Option to display OCR'ed text #325

New issue

Closed

opened 2026-02-25 21:31:42 +03:00 by kerem · 2 comments

kerem commented

2026-02-25 21:31:42 +03:00

Owner

Originally created by @eric-saintetienne on GitHub (Oct 24, 2021).
Original GitHub issue: https://github.com/ciur/papermerge/issues/424

Originally assigned to: @ciur on GitHub.

As of 2.0, the OCR text is displayed on top of the image, as an overlay.
Displaying the OCR'ed text in a text area would be useful, at least to check that OCR worked as expected (not many recognition errors) but also to select/copy the OCR'ed text to the clipboard for saving.

For the implementation, it's up to you, I think a textarea could be added under the metadata, on the right hand side of a file, or it could be a separate page containing a textarea and accessed via a new entry in the contextual (right click) menu for a given document.

The alternative is to open the image viewer and do Ctrl+A, but there are problems:

Ctrl+A selects the whole page, including a bunch of other text (whatever is displayed on the page, like the html div and spans)
Once selected with Cltr+A the overlayed OCR'ed text is still hard to read (it's not as readable as a textearea)

Thanks!

Originally created by @eric-saintetienne on GitHub (Oct 24, 2021). Original GitHub issue: https://github.com/ciur/papermerge/issues/424 Originally assigned to: @ciur on GitHub. As of 2.0, the OCR text is displayed on top of the image, as an overlay. Displaying the OCR'ed text in a text area would be useful, at least to check that OCR worked as expected (not many recognition errors) but also to select/copy the OCR'ed text to the clipboard for saving. For the implementation, it's up to you, I think a textarea could be added under the metadata, on the right hand side of a file, or it could be a separate page containing a textarea and accessed via a new entry in the contextual (right click) menu for a given document. The alternative is to open the image viewer and do Ctrl+A, but there are problems: 1. Ctrl+A selects the whole page, including a bunch of other text (whatever is displayed on the page, like the html div and spans) 2. Once selected with Cltr+A the overlayed OCR'ed text is still hard to read (it's not as readable as a textearea) Thanks!

kerem

2026-02-25 21:31:42 +03:00

closed this issue
added the
enhancement

feature request
labels

kerem commented

2026-02-25 21:31:42 +03:00

Author

Owner

@ajarzyn commented on GitHub (Nov 29, 2021):

Hello @eric-saintetienne,

You already can do this, but I must admit that this option is rather hidden, so maybe good suggestion would be to make it more accessible.

How to view OCRed text of the page:

Open your document
Select this icon in the left top corner:
Select page you would like to see OCR from (this is important without selection option won't work)
Press right mouse button on the page
Select "View OCRed text"

I may make a pull request to add this description to documentation.
@ciur would that be a good idea?

@ajarzyn commented on GitHub (Nov 29, 2021): Hello @eric-saintetienne, You already can do this, but I must admit that this option is rather hidden, so maybe good suggestion would be to make it more accessible. How to view OCRed text of the page: 1. Open your document 2. Select this icon in the left top corner: ![obraz](https://user-images.githubusercontent.com/18590439/143952224-4689c9d7-d13f-47f7-80d5-990d8eeaca3a.png) 3. Select page you would like to see OCR from (this is important without selection option won't work) 4. Press right mouse button on the page 5. Select "View OCRed text" I may make a pull request to add this description to documentation. @ciur would that be a good idea?

kerem commented

2026-02-25 21:31:42 +03:00

Author

Owner

@ciur commented on GitHub (Aug 27, 2022):

Feature is available in 2.1.0x and it is more intuitive to use.
Here is a quick demos:

@ciur commented on GitHub (Aug 27, 2022): Feature is available in 2.1.0x and it is more intuitive to use. Here is a quick demos: ![select-first-page-is-not-necessary-anymore](https://user-images.githubusercontent.com/24827601/187017685-ee64bcde-e72a-4d1b-bea6-8929014f6a4f.gif)