New public dataset for spotting patterns in medieval document images.
Sovann EnStéphane NicolasCaroline PetitjeanFrédéric JurieLaurent HeuttePublished in: J. Electronic Imaging (2017)
Keyphrases
- document images
- document analysis
- document image analysis
- document image understanding
- document image retrieval
- optical character recognition
- page segmentation
- word level
- word spotting
- printed documents
- page layout
- line extraction
- document processing
- language identification
- scanned documents
- historical documents
- hidden markov models
- mathematical formulas
- comparative evaluation
- printed text
- image set
- level set