Open-source OCR / ATR system for historical documents and handwritten materials.
Kraken OCR is an open-source text recognition system particularly suited to historical documents, non-Latin script materials and handwriting recognition tasks that need custom training. It supports layout analysis and model training, making it suitable for humanities and social sciences, archive digitisation and historical document research.
Use cases: historical document OCR, handwriting recognition, archive digitisation, model training, humanities and social science material organisation
Note: Configuration and training are more demanding than ordinary OCR software, so it suits users with some Python or command-line experience.
