Request for a Good Arabic OCR Model for Tesseract

[Abdlrhman00]

Your Feature Request

Hello everyone,

I hope you’re all doing well. I’ve been working on improving Arabic OCR using Tesseract, but I’ve struggled to achieve high accuracy. Training a model from scratch has been challenging, and I haven’t been able to get satisfactory results so far.

If anyone has already trained a good Arabic model and is willing to share their .traineddata file, it would mean a lot to me. I believe sharing these resources could benefit many of us working on similar projects.

I would truly appreciate any help, and if there’s anything I can contribute in return, I’d be happy to do so.

Thank you in advance, and looking forward to hearing from you!

Best regards,