Automated OCR Ground Truth Generation.
Joost van BeusekomFaisal ShafaitThomas M. BreuelPublished in: Document Analysis Systems (2008)
Keyphrases
- ground truth
- post processing
- high quality
- optical character recognition
- semi automated
- character recognition
- ground truth data
- preprocessing
- data mining
- automated analysis
- gold standard
- machine learning
- semi automatic
- document images
- context sensitive
- manually labeled
- text recognition
- neural network
- generation process
- printed documents
- handwriting recognition
- character segmentation
- document image retrieval
- segmented images
- human subjects
- test images
- data driven
- medical images
- data structure
- e learning