I finetuned the tesseract for a custom font and added symbols. it works okayish with psm 7 on single line text (it hallucinates sometimes. it has trouble detecting '.' but I am trying to fix that). However it messed up the characters ,i.e. adds some or inserts random if there are multiple lines that aren't aligned (subscript and superscript). The exact same individual lines work fine in psm 7. I tried with all psm modes. Am i doing something wrong or is it an limitation of the model? Another issue with rotated text. Some text blocks are at random rotation, it doesn't work (outputs gibberish), again tried all psm modes. What could be the possible causes and how can i fix them ?
-- 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 visit https://groups.google.com/d/msgid/tesseract-ocr/1e00527a-e13e-4fe6-9671-4a64844660cbn%40googlegroups.com.