Hello, I just generated the traineddata file for an old historical version of latin text, but when I run tesseract on the .tif that I used to train tesseract for the language (as well as with other sample images), it returns an empty result. However, when I use the English language for classification, it generates text with a few errors due to a lack of recognition for some specific characters. (Meaning that the fault lies with the traineddata and not the samples I am running it on)
Why could this be? I have been struggling to even generate the traineddata, and ended up using a fairly short training text (see attachment). Do I need to use a longer training text/tif? If anyone could point me in the right direction I would be extremely grateful. Thanks in advance! -Brennan -- 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/29355c0a-deeb-4f65-a176-9abae60bcb9c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
nlg.traineddata
Description: Binary data
