A New Strategy for Arabic OCR: Archigraphemes, Letter Blocks, Script Grammar, and shape synthesis.
Thomas MiloAlicia González MartínezPublished in: DATeCH (2019)
Keyphrases
- optical character recognition
- printed documents
- character recognition
- language identification
- document images
- preprocessing
- post processing
- shape model
- texture synthesis
- document processing
- handwriting recognition
- shape analysis
- error correction
- text recognition
- shape descriptors
- natural language
- formal languages
- fractal image coding
- ocr systems
- recognition errors
- grammatical inference
- medial axis
- shape representation
- shape features
- machine vision
- hidden markov models