Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

problem with arabic lines #1338

Open
zMuh opened this issue Nov 23, 2024 · 3 comments
Open

problem with arabic lines #1338

zMuh opened this issue Nov 23, 2024 · 3 comments

Comments

@zMuh
Copy link

zMuh commented Nov 23, 2024

when i try to pass an image with arabic text idont get lines detected correctly can i get any help ?

Output

@romanvelichkin
Copy link

romanvelichkin commented Dec 9, 2024

If you need horizontal lines to be detected, you have to fine-tune CRAFT model by yourself.

https://github.com/JaidedAI/EasyOCR/tree/master/trainer/craft

@rezaies
Copy link

rezaies commented Dec 12, 2024

Hi @romanvelichkin,

I think you didn't understand what the original reporter was trying to say.
Take a look at the second line in the text, for example. I mean the line after the short line. You can see that six blocks (I'm not familiar with OCR and don't know what they are called) are detected in a single line. Each of these blocks will appear as a separate line in the output.
To make matters worse, they are printed one by one, from left to right, in a text that is supposed to be written from right to left!

@romanvelichkin
Copy link

Hi @rezaies,
So he meant lines of text, I thought he meant that little line at the bottom of page.

To make the boxes more sparse, there are many settings that can be set for this before scanning.

To make matters worse, they are printed one by one, from left to right, in a text that is supposed to be written from right to left!

I think it is not that hard to write a parser for the scan results, given this situation.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants