Volltexterkennung für historische Sammlungen mit OCR4all-libraries iterativ und partizipativ gestalten.
Jan Sebastian KlaesKristof KorwisiKatharina KrügerChristian ReulNadine TowaraPublished in: DHd (2022)
Keyphrases
- optical character recognition
- digital libraries
- character recognition
- post processing
- document images
- test set
- preprocessing
- error correction
- document image analysis
- text recognition
- recognition errors
- document processing
- massachusetts institute of technology
- real time
- ocr systems
- printed documents
- handwriting recognition
- image segmentation
- website
- artificial intelligence
- machine learning