Re:OCR and Medieval Manuscripts: Establishing a Baseline (2015)
by mdani on Thursday November 16, 2017 @03:22PM
(#15716236)
In my experience, tesserect 4 with LSTM is far more accurate in recognizing characters. It was not there in 2015, so the post would probably need an update or a follow up.
https://github.com/tesseract-ocr/tesseract/wiki/4.0-with-LST...