Title :
Handwritten Character Recognition Using HMM Model Based on Bagging Method
Author :
Zhang, Yan ; Yao, Xiaodong ; Chang, Ching
Author_Institution :
Dept. of Electron. Inf., East China Univ. of Sci. & Technol., Shanghai, China
Abstract :
The purpose of this paper is to improve recognition rate of off-line handwritten character recognition system.We apply the statistical characteristics of the percentage of pixels and structural characteristics of boundary chain code of character projection, after train based on HMM to obtain corresponding parameters, then integrate different classifiers through the Bagging algorithm in Voting method. Experimental results indicate that this approach can further improve the performance.
Keywords :
handwriting recognition; handwritten character recognition; hidden Markov models; Bagging algorithm; Bagging method; Voting method; boundary chain code; character projection; handwritten character recognition rate; hidden Markov models; offline handwritten character recognition system; Bagging; Character recognition; Data mining; Data preprocessing; Feature extraction; Handwriting recognition; Hidden Markov models; Image recognition; Image segmentation; Pattern recognition;
Conference_Titel :
Computational Intelligence and Software Engineering, 2009. CiSE 2009. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-4507-3
Electronic_ISBN :
978-1-4244-4507-3
DOI :
10.1109/CISE.2009.5366197
