I tried manually adding files I needed from https://github.com/tesseract-ocr/tessdata_best (`equ.traineddata`, `osd.traineddata`, `ita.traineddata`) inside `/usr/share/tesseract-ocr/5/tessdata`: unfortunately I then found out the hard way that these only work on Tesseract 4 XD.
1. It seems funny though: does that really mean I'll get better results by downgrading so that I can actually use these files? I understand the performance loss, but I'm particularly interested in getting the best of `equ.traineddata`, which to my understanding interprets math characters, which are often a challenge for OCR engines, so was trying to get the absolute best scan possible for that. 2. Also, I wasn't able to specify `-l equ` as the error told me Tesseract is supposed to deal with that on its own: if that's the case, is `equ` installed by default with `sudo apt-get install tesseract-ocr` (couldn't find it in `tessdata` folder, and don't know where else to look for it)? 3. I also tested the Docker image: if I put `equ.traineddata` and `osd.traineddata` inside the `tessdata` folder will they (which I have chosen manually) actually be used? Hope this all makes sense, don't be afraid to ask :) Alessandro -- 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 view this discussion visit https://groups.google.com/d/msgid/tesseract-ocr/789d7514-bded-49e4-95ed-44cfb0049ad1n%40googlegroups.com.
