1. Implement text detection on the image (EAST, YOLO... see
https://www.youtube.com/watch?v=ZpRNfWzuexQ) or search for "text
detection python"
2. Process detected areas so there is a text without any graphics - see
some suggestions in docs (
https://github.com/tesseract-ocr/tessdoc/blob/main/ImproveQuality.md)
3. run OCR (tesseract) on the processed area(s)
Zdenko
pi 9. 12. 2022 o 18:15 Anna Muller <amull...@nd.edu> napísal(a):
> Hello - I am very new to using the Tesseract software. I am currently
> completing a project that requires me to read text from TikTok screenshots
> - I attached a random example image I got from Tiktok to this post. I am
> currently getting pretty inaccurate output. Below I pasted the code I was
> using that I got from an online tutorial.
>
> I was wondering if anybody had any suggestions or would be able to point
> me in the right direction to resources where I could better learn how to
> fine tune my image processing parameters.
>
> I am currently using Jupyter Notebooks, but if anybody suggests accessing
> Tesseract differently, please let me know.
>
> *My Code:*
> from PIL import Image
> column = Image.open('tiktoktest.png')
> gray = column.convert('L')
> blackwhite = gray.point(lambda x: 0 if x < 200 else 255, '1')
> blackwhite.save("tiktok.jpg")
>
> text_from_image = pytesseract.image_to_string(Image.open('tiktok.jpg'))
> print(text_from_image)
>
> --
> 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-ocr+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/tesseract-ocr/e7c6a467-e53d-442d-b7bf-1fd645cdd66an%40googlegroups.com
> <https://groups.google.com/d/msgid/tesseract-ocr/e7c6a467-e53d-442d-b7bf-1fd645cdd66an%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
--
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-ocr+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/tesseract-ocr/CAJbzG8yOcfPKssSd722zkfyQFZWx1fEx379iwGvi7bP132h6qA%40mail.gmail.com.
