Vom gedruckten Werk zu elektronischem Volltext als Forschungsgrundlage Erstellung von Forschungsdaten mit OCR-Verfahren.
Matthias BoenigElisa HerrmannVolker HartmannPublished in: DHd (2019)
Keyphrases
- optical character recognition
- document images
- character recognition
- post processing
- error correction
- recognition errors
- test set
- preprocessing
- text recognition
- massachusetts institute of technology
- scanned documents
- printed documents
- document processing
- machine vision
- neural network
- computer vision
- document image analysis
- end to end
- training set
- multiscale
- face recognition