<https://lh3.googleusercontent.com/-Ih4uQKB9hpc/V9eUe7ddQsI/AAAAAAAAAAM/skGFAMee73smMx4Q9U2NWZJ62oF0GrFAwCLcB/s1600/MenuAllText.png> I just started using tesseract today with PyTesseract. I'm trying to have it read text from a game, it seems easy enough but I'm completely stuck. The picture is above is one I was trying to read. I thought it was a good test since there is just a black background and nothing to really mistake for a character. I can't get this to read at all though. If I scan in English it doesn't recognize Japanese. If I scan in Japanese it inserts kanji in place of English letters. If I use both it still misses a bunch of writing. I've tried just cropping down to the dialog box, but then it only reads the last 2 lines. If I shrink the image it'll only read the first half of the top two lines. Changing to gray-scale did nothing, forcing it to full contrast (either black or white pixels) made it worse. Also, when just scanning the dialog box it got basically every kanji wrong, it skipped other characters entirely, it also reads ら as a 6. Is this a font issue? Or does anyone know if there's some tricks to help with this? I'd appreciate any suggestions since I can't seem to get it to go anywhere.
-- 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/0fd15456-ee1b-4427-8e5c-667db9b2b71d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
