I don't know if the Tess4J wrapper supports multiple languages, or which language you're using as a base language, but you might consider training it and any other symbols you need into an entirely separate language and then OCRing using the Tess4J equivalent of -l eng+mylang (or whatever your base language is).
There is the equation "language" with code equ, but it apparently doesn't include that style division sign, which I was a little surprised at. You might try -l eng+equ to start though and see perhaps the "wavy division sign" or some other symbol is close enough that it gets detected reliable in place of the plus sign, before going to the trouble of doing your own training. I've attached the characters which are included in the equ training. Tom On Saturday, February 13, 2016 at 5:13:46 PM UTC-5, Quan Nguyen wrote: > > You'd train Tesseract and then use the resultant .trainneddata file with > Tess4J. > > https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract > > On Saturday, February 13, 2016 at 7:38:24 AM UTC-6, Alex wrote: >> >> How would I go about training for new Unicode symbols for Tess4J. I need >> Tesseract to detect a division symbol (รท), but it detects it as a plus >> sign. >> >> Thanks. >> > -- 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 [email protected]. To post to this group, send email to [email protected]. 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/325d2dab-b81f-46dd-bdf4-b3c4bfa5b921%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
equtmp.unicharset
Description: Binary data
