Closed CypherousSkies closed 2 years ago
Here's a post that should help fix bad ocr https://medium.com/states-title/using-nlp-bert-to-improve-ocr-accuracy-385c98ae174c
splitting remove page numbers to #9 as its going to take substantial work.
functionally ocrmypdf is implemented now, modulo opencv tasks (#9) which is beyond my scope for short term work
Todo: