I run a standard Ubuntu Linux and now this means Tesseract 5.1.0.
Back under Tesseract 4.0 and 4.1, there were two separate files swe-frak.traineddata and dan_frak.traineddata that one could download and get very good results with old Swedish and Danish book pages in Gothic/Fraktur typesetting. When I upgraded Ubuntu Linux and got Tesseract 5.1, these were missing. When I copy the old file from /usr/share/tesseract-ocr/4.00/tessdata/dan_frak.traineddata to /usr/share/tesseract-ocr/5/tessdata/dan_frak.traineddata It doesn't produce very useful results. There are lots of errors, far more than expected. What to do? Should those, who made those files, make new versions that will work with the new Tesseract? Or will Tesseract finally incorporate Fraktur reading without the need to load separate training files? -- Lars Aronsson (l...@aronsson.se) Project Runeberg - free Nordic literature - http://runeberg.org/ -- 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/859aba88-e90c-e78b-19c1-1d7a851bce95%40aronsson.se.
