Off-line Character Recognition Using HMM by Multiple Directional Feature Extraction and Voting with Bagging Algorithm, Proceedings of 5th International Conference on Document Analysis and Recognition (ICDAR'99), pp. 49-52 (1999)
Abstract
The purpose of our research is to improve the recognition rate of off-line character recognition systems using the HMM (Hidden Markov Model) without increasing a number of HMM parameters too much. Some 2-dimensional HMM character recognition systems have been proposed to increase representational power. However, since 2-D HMM has much more complex structure than 1-dimensional HMM, it becomes very hard to gather sufficient samples in order to guarantee the successful generalization. To overcome the problem, we propose a method for character recognition using 1-D HMMs in multiple directions with 2-dimensional feature extraction. To further improve the performance, some voting methods using bagging algorithm are also exploited. In our experiment, the recognition rate is increased by 1% with the multiple directional HMM character recognition system compared to the 1-D HMM character recognition system. The recognition rate is further increased by about 1% with the HMM character recognition system using bagging algorithm.
←[手書き文字認識]
←[中野の研究紹介]
←[中野の目次]
| mail address: | |
← お手数ですが打ち込んで下さい |
First Written March 23, 2000
Transplanted to KSU Before May 15, 2003
Transplanted to So-net May 3, 2005
Last Update April 8, 2007
© Yasuaki Nakano 2000-2007