HMM-based script identification for OCR.
Dmitriy GenzelAshok C. PopatRemco TeunenYasuhisa FujiiPublished in: MOCR@ICDAR (2013)
Keyphrases
- optical character recognition
- hidden markov models
- text recognition
- post processing
- document images
- character recognition
- recognition errors
- error correction
- preprocessing
- handwriting recognition
- printed documents
- data sets
- scanned documents
- variable duration
- page layout
- character segmentation
- document processing
- viterbi algorithm
- machine vision
- document image analysis
- document analysis
- multimedia
- database