| Wavelet Neural Network for Handwritten Manchu Character Unit Recognition |
| |
| 引用本文： | ZHANG Guangyuan LI Jingjiao WANG Aixia. Wavelet Neural Network for Handwritten Manchu Character Unit Recognition[J]. 现代电子技术, 2007, 30(1): 118-120 |
| |
| 作者姓名： | ZHANG Guangyuan LI Jingjiao WANG Aixia |
| |
| 作者单位： | [1]School of Information Science ＆ Engineering, Shenyang University, Shenyang, 110044, China [2]School of Information Science ＆ Engineering, Northeastern University, Shenyang, 110004, China |
| |
| 摘 要： | The Manchu character recognition method based on Manchu character unit is an efficient method. In this method, the recognition accuracy rate of Manchu character unit has great influence on the final recognition result. As new approach to solve this problem,a hybrid wavelet neural network scheme has been developed as an assistant method combine with the original combo - distance method. Due to the properties of the wavelet neural network,the training problem can be transformed into a convex optimization process,therefore the global minimum can be obtained and the learning speed is increases. Both the learning samples set and testing samples set are used, experimental results demonstrate the combine method based on the wavelet neural network is more efficient than the single combo - distance method.
|
| 关 键 词： | 小波神经网络 字符单元 满族字符识别法 模式识别 |
| 文章编号： | 1004-373X(2007)01-118-03 |
| 修稿时间： | 2006-06-16 |
Wavelet Neural Network for Handwritten Manchu Character Unit Recognition |
| |
| ZHANG Guangyuan,LI Jingjiao,WANG Aixia. Wavelet Neural Network for Handwritten Manchu Character Unit Recognition[J]. Modern Electronic Technique, 2007, 30(1): 118-120 |
| |
| Authors: | ZHANG Guangyuan LI Jingjiao WANG Aixia |
| |
| Abstract: | The Manchu character recognition method based on Manchu character unit is an efficient method.In this method,the recognition accuracy rate of Manchu character unit has great influence on the final recognition result.As new approach to solve this problem,a hybrid wavelet neural network scheme has been developed as an assistant method combine with the original combo-distance method.Due to the properties of the wavelet neural network,the training problem can be transformed into a convex optimization process,therefore the global minimum can be obtained and the learning speed is increases.Both the learning samples set and testing samples set are used,experimental results demonstrate the combine method based on the wavelet neural network is more efficient than the single combo-distance method. |
| |
| Keywords: | wavelet neural network Manchu character recognition |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
