Train Tesseract to ignore music?

[Sara Palmer]

I'd like to produce high-quality OCR of books that contain text interspersed with music. Is it possible to train Tesseract to ignore musical notation instead of turning it into junk OCR? How would one go about doing this?

[Lorenzo Bolzani]

Hi Sara, can you please post a sample picture? You could probably detect the pentagram (hough lines with very tight paramters, custom horizontal lines detection) and just replace it with a white rectangle.

Lorenzo

Il giorno ven 28 giu 2019 alle ore 07:15 Sara Palmer <sara....@gmail.com> ha scritto:

I'd like to produce high-quality OCR of books that contain text interspersed with music. Is it possible to train Tesseract to ignore musical notation instead of turning it into junk OCR? How would one go about doing this?

--
You received this message because you are subscribed to the Google Groups "tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-oc...@googlegroups.com.
To post to this group, send email to tesser...@googlegroups.com.
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/3988b3d7-d757-4c10-9a66-f7aa34a65b6f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Timothy Snyder]

A picture would be helpful. From my experience, however, writing an independent program to segment text from "noisy" images with a lot of non-text print will give you the best results. Depending on how much the layout of those books varies between pages, this could be a simple or complicated task.

To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CAMgOLLy4JJnVwrkwZaTsUhb5r3XTETaWcy_wmDbQdGQJv9t4XA%40mail.gmail.com.

[Sara Palmer]

Thank you both for your suggestions! I've attached a couple sample images.