Off-Line, Handwritten Numeral Recognition by Perturbation Method
IEEE Transactions on Pattern Analysis and Machine Intelligence
Parsing and Recognition of City, State, and ZIP Codes in Handwritten Addresses
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
A2iA Check Reader: A Family of Bank Check Recognition Systems
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
IWFHR '02 Proceedings of the Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)
Hidden Markov Model Length Optimization for Handwriting Recognition Systems
IWFHR '02 Proceedings of the Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)
IWFHR '02 Proceedings of the Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
Offline Recognition of Large Vocabulary Cursive Handwritten Text
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 2
Recognition of Cursive Roman Handwriting - Past, Present and Future
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
Offline Grammar-Based Recognition of Handwritten Sentences
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
This paper investigates the application of a probabilistic parser for natural language on the ist of the N-best sentences produced by an off-ine recognition system for cursive handwritten sentences. For the generation of the N-best sentence ist an HMM-based recognizer including a bigram anguage model is used. Theparsing of the sentences is achieved by a bottom-upchart parser for stochastic context-free grammars whichproduces the parse tree of the input sentence as well asthe word tags. From a collection of corpora we extractthe linguistic resources to build the lexicon, a word bigram model and the stochastic context-free grammar.Results from experiments indicate an increase of theword and sentence recognition rate when using the proposed combination scheme.