Evaluation of Different Strategies to Optimize an HMM-Based Character Recognition System

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 HMMbased recognition method are very promising.

• Type: Journal
• Year: 2009
• Where: ICDAR
• Authors: Murilo Santos, Albert Hung-Ren Ko, Luiz S. Oliveira, Robert Sabourin, Alessandro L. Koerich, Alceu de Souza Britto Jr.
• Keywords: Document Analysis, ICDAR 2009, Noise Reduction, Noise Reduction Method, Noise Reduction Process