Paper Info
Title
Evaluation of Different Strategies to Optimize an HMM-Based Character Recognition System
Abstract
Different strategies for combination of complementary features in an HMM-based method for handwritten character recognition are evaluated. In addition, a noise reduction method is proposed to deal with the negative impact of low probability symbols in the training database. New sequences of observations are generated based on the original ones, but considering a noise reduction process. The experimental results based on 52 classes of alphabetic characters and more than 23,000 samples have shown that the strategies proposed to optimize the HMM-based recognition method are very promising.
Year
DOI
Venue
2009
10.1109/ICDAR.2009.230
ICDAR-1
Keywords
Field
DocType
low probability symbol,different strategy,hmm-based recognition method,hmm-based character recognition system,different strategies,noise reduction method,handwritten character recognition,noise reduction process,alphabetic character,complementary feature,hmm-based method,feature extraction,text analysis,pixel,handwriting recognition,neural networks,stochastic processes,noise reduction,probability,hidden markov models,optimization,noise
Noise reduction,Character recognition,Pattern recognition,Computer science,Handwriting recognition,Speech recognition,Feature extraction,Image denoising,Pixel,Artificial intelligence,Hidden Markov model
Conference
Citations 
PageRank 
References 
1
0.40
9
Authors
6
Name
Order
Citations
PageRank
Murilo Santos110.40
Albert Ko210.40
Luis S. Oliveira350.84
robert sabourin4109573.81
Alessandro L. Koerich552539.59
Alceu Britto69418.30