Hello Everyone, Where can I find the box/tif combo for the eng.traineddata that Tessearct 3.02 provides for download?
Since I cannot append new font information to compiled training data I have to completely create new data even for the fonts that Tesseract's prebuilt training provides. When I do this I lose accuracy over the provided training data significantly (accuracy ends up being around 50% down from above 90%). If I had Tessearct's source box/tif files then adding my fonts should at worst still have nearly the same accuracy as the project provided files for documents that contain the default fonts. It would seem that Tesseract providing the box/tif originally used would be the easiest solution in keeping user's accuracy up while attempting new fonts. I have found box/tif files for tessearct 2.0 but not 3.0. When I use the box/tif files from 2.0 for the fonts provided like Arial, Courier New, etc I significantly lose accuracy. -- 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 http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/88f08cfe-ef4a-4b20-8b95-1bdb29b702b4%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
