Recognition of Broken Characters from Historical Printed Books Using Dynamic Bayesian Networks.
Laurence Likforman-SulemMarc SigellePublished in: ICDAR (2007)
Keyphrases
- dynamic bayesian networks
- optical character recognition
- historical manuscripts
- bayesian networks
- approximate inference
- particle filtering
- object recognition
- structure learning
- feature extraction
- character recognition
- probability distribution
- graphical models
- visual tracking
- gaussian process
- state variables
- support vector
- handwriting recognition