erl-ang / interactive-ocr

Implementation of a couple of heuristics that estimate OCR quality without reliance on ground truth data, focusing on historical documents written in English.

Geek Repo

Github PK Tool

ground-truth nlp ocr-quality optical-character-recognition tesseract-ocr word-error-rate

interactive-ocr

cleaning this up is on my postgrad todo list

About

Implementation of a couple of heuristics that estimate OCR quality without reliance on ground truth data, focusing on historical documents written in English.

ground-truth nlp ocr-quality optical-character-recognition tesseract-ocr word-error-rate

Languages

Language:Python 95.8%Language:Shell 3.5%Language:Ruby 0.7%