| 基于Bayes决策的手写体数字识别 |
| |
| 引用本文： | 章慎锋,杨淑莹,王厚雪. 基于Bayes决策的手写体数字识别[J]. 天津理工大学学报, 2006, 22(1): 78-80 |
| |
| 作者姓名： | 章慎锋 杨淑莹 王厚雪 |
| |
| 作者单位： | 天津理工大学,计算机科学与工程系,天津,300191 |
| |
| 摘 要： | 提出了一种用Bayes决策理论进行手写体数字识别的方法，对已知类别的样品提取特征建立数字样品库，对于任意的手写数字提取特征，根据数字样品库中已知样品的特征，运用基于最小错误概率的Bayes决策进行识别、实验证明Bayes决策理论用于手写体数字的识别有较好的效果，一般情况下识别率能达到96％以上．
|
| 关 键 词： | Bayes决策 最小错误概率 模式识别 手写体数字 |
| 文章编号： | 1673-095X（2006）01-0078-03 |
| 收稿时间： | 2005-07-11 |
| 修稿时间： | 2005-07-11 |
Handwritten number recognition based on Bayes decision |
| |
| ZHANG Shen-feng,YANG Shu-ying,WANG Hou-xue. Handwritten number recognition based on Bayes decision[J]. Journal of Tianjin University of Technology, 2006, 22(1): 78-80 |
| |
| Authors: | ZHANG Shen-feng YANG Shu-ying WANG Hou-xue |
| |
| Affiliation: | Department of Computer Science and Engineering, Tianjn University of Technology, Tianjin 300191, China |
| |
| Abstract: | A new scheme for handwritten number recognition is proposed in this paper. According to features of samples whose classifications are known, a sample room is set up. To recognize a handwritten number, we should extract its features. Combining with the samples in sample room, its classification can be confirmed via Bayes decision based on the least error probability. Experiments show that the proposed scheme can classify handwritten number well, whose recognition rate is above 96%. |
| |
| Keywords: | Bayes decision least error probability pattern recognition handwritten number |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
