> > That's interesting that you tried replacing the top layer. I haven't > tried that yet. How many iterations did you use? > >> In this case the unicharset was limited to UPPERCASE letters, 0-9 numbers , : and /. I used a training_text which followed the pattern of the image - lines starting with LOT# and EXP: and using similar pattern. I used 2 fonts which were very similar to the image. So this was narrowly focussed on single use and only 2000 iterations were needed with tessdata_best/eng to get error rate down to 0.2 or so. The # of iterations for plus training were also similar but they did not give same accuracy (also, the traineddata file size is much smaller using this method).
-- 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/CAG2NduVSCJMvs7OCgQhfS%3D_V0yURmiJ%2BNSD9bb6ZGaoN6Tt8Pg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
