Hi Zdenop, thank you for your reply. The input images are dynamic, can be multiline or singleline. I attached only image which i have problem. Yes psm 7 the example image works fine somehow. In multiline context (psm 6) it recognizes hyphens fine when it together with other characters, but does not recognize single hyphen when alone. I wonder is there any way to train to recognize correctly single hyphen. Or any tesseract option.
Thanks, Ankhbayar On Sunday, October 6, 2024 at 1:57:30 AM UTC+8 zdenop wrote: > I’m having trouble understanding the issue you're encountering—you > mentioned that your input could be multiline, but the example you provided > only contains a single character. I’m also surprised that even *psm 7* > works in this case. > > The problem is that a hyphen, when presented as a single character, can > sometimes be interpreted as a vertical line or graphical element and may be > ignored. Typically, in a multiline context, it's recognized correctly. > > Zdenko > > > pi 4. 10. 2024 o 19:16 'Ankhbayar Gansukh' via tesseract-ocr < > tesser...@googlegroups.com> napísal(a): > >> Hello, >> I am new to tesseract and would be very appreciate if someone help me. >> >> Tesseract5 recognizes succesfully single hyphen in psm=7. >> Example: tesseract --psm 7 single_hyphen_char.jpg stdout >> >> When use psm=6 it does not recognize single hyphen. >> Example: tesseract --psm 6 single_hyphen_char.jpg stdout >> >> I cannot use psm 7 because image can be multiline. >> >> Any advice would be greatly appreciated! >> Ankhbayar >> >> >> >> >> >> -- >> 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 view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/ab72fe8c-61d5-4a6f-936b-69e9658c03a7n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/ab72fe8c-61d5-4a6f-936b-69e9658c03a7n%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/4a02a255-786e-4fa9-b3fc-2b5bcd339824n%40googlegroups.com.
