95 articles 

inria-00104839, version 1

Modeling of Pen-Coordinate Information in SCPR-based HMM for On-line Recognition of Handwritten Japanese Characters

Junko Tokuno () 1, Yiping Yang () 1, Gleidson Pegoretti Da Silva () 1, Akihito Kitadai () 1, Masaki Nakagawa () 1

Tenth International Workshop on Frontiers in Handwriting Recognition (2006)

Abstract: This paper describes stochastic modeling of pencoordinate information in HMMs with structured character pattern representation (SCPR) for on-line recognition of handwritten Japanese characters. SCPR allows HMMs for Kanji character patterns to share common subpatterns. Although SCPR-based HMMs have been successfully applied to Kanji character recognition, the pen-coordinate feature has not been modeled since it is an unique feature for each character pattern. In this paper, we employ mapping from a common subpattern to each occurrence in Kanji patterns in the estimation step of SCPR-based HMMs. We also employ adaptation of state parameters to each character pattern in generating character HMMs by composing SCPR-based HMMs. Experimental results showed that the correct recognition rate improved from 83.6% to 92.3% for Japanese Kanji character patterns by modeling the pen-coordinate information in the SCPRbased HMMs.

  • 1:  Tokyo University of Agriculture and Technology
  • Tokyo University of Agriculture and Technology
  • Domain : Computer Science/Document and Text Processing
    Computer Science/Computer Vision and Pattern Recognition
  • Keywords : On-line handwriting recognition – Structured character pattern representation – HMM – Pen-coordinate feature
  • Comment : http://www.suvisoft.com
 
  • inria-00104839, version 1
  • oai:hal.inria.fr:inria-00104839
  • From: 
  • Submitted on: Monday, 9 October 2006 15:08:03
  • Updated on: Monday, 9 October 2006 15:11:54