Title :
Optimizing the Number of States for HMM-Based On-line Handwritten Whiteboard Recognition
Author :
Geiger, Jürgen ; Schenk, Joachim ; Wallhoff, Frank ; Rigoll, Gerhard
Author_Institution :
Inst. for Human-Machine Commun., Tech. Univ. Munchen, Munich, Germany
Abstract :
In this paper, we present a novel way to determine the number of states in Hidden-Markov-Models for on-line handwriting recognition. This method extends the Bakis length modeling method which has successfully been applied to off-line handwriting recognition. We propose a modification to the Bakis method and present a technique to improve the topology with a small number of iterations. Furthermore, we investigate the influence of state tying. In an experimental section, we show that our improved system outperforms a system with Bakis length modeling by 1.5 % relative and with fixed length modeling by 5.1 % relative on the IAM-On-DB-t1 benchmark.
Keywords :
handwriting recognition; handwritten character recognition; hidden Markov models; Bakis length modeling method; HMM-based online handwritten whiteboard recognition; hidden Markov model; offline handwriting recognition; online handwriting recognition; HMM Topology; Handwriting Recognition; Number of HMM states;
Conference_Titel :
Frontiers in Handwriting Recognition (ICFHR), 2010 International Conference on
Conference_Location :
Kolkata
Print_ISBN :
978-1-4244-8353-2
DOI :
10.1109/ICFHR.2010.23