OCR-D & OCR4all: Two Complementary Approaches for Improved OCR of Historical Sources.
Konstantin BaiererAndreas BüttnerElisabeth EnglLena HinrichsenChristian ReulPublished in: HistoInformatics@JCDL (2021)
Keyphrases
- optical character recognition
- document images
- post processing
- recognition errors
- character recognition
- error correction
- preprocessing
- text recognition
- scanned documents
- printed documents
- ocr systems
- document image analysis
- document image retrieval
- document processing
- document analysis
- databases
- decision trees
- image processing
- page layout
- machine learning algorithms
- text extraction
- e learning
- real time