23 January 2012 Construction of language models for an handwritten mail reading system
Author Affiliations +
This paper presents a system for the recognition of unconstrained handwritten mails. The main part of this system is an HMM recognizer which uses trigraphs to model contextual information. This recognition system does not require any segmentation into words or characters and directly works at line level. To take into account linguistic information and enhance performance, a language model is introduced. This language model is based on bigrams and built from training document transcriptions only. Different experiments with various vocabulary sizes and language models have been conducted. Word Error Rate and Perplexity values are compared to show the interest of specific language models, fit to handwritten mail recognition task.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Olivier Morillot, Olivier Morillot, Laurence Likforman-Sulem, Laurence Likforman-Sulem, Emmanuèle Grosicki, Emmanuèle Grosicki, "Construction of language models for an handwritten mail reading system", Proc. SPIE 8297, Document Recognition and Retrieval XIX, 82970S (23 January 2012); doi: 10.1117/12.911965; https://doi.org/10.1117/12.911965

Back to Top