| 基于MHMM模型的手写体汉字识别算法 |
| |
| 引用本文： | 刘健,李会方,牛新伟. 基于MHMM模型的手写体汉字识别算法[J]. 信息安全与通信保密, 2007, 0(2): 75-77 |
| |
| 作者姓名： | 刘健 李会方 牛新伟 |
| |
| 作者单位： | 西北工业大学电子信息学院,陕西,西安,710072 |
| |
| 摘 要： | 文章提出了一种基于多重隐马尔可夫模型和区域投影变换的手写体汉字识别新方法。该方法对每个汉字建立4个HMM,通过等比重综合方法将4个分类器的计算结果进行综合,从而得到识别结果。该方法对于识别应用问题中的隐马尔可夫建模具有普适性。实验证明,该识别方法具有较高的识别准确率。
|
| 关 键 词： | 汉字识别 特征提取 多重隐马尔可夫模型 边界链编码 |
| 文章编号： | 1009-8054(2007)02-0075-03 |
| 修稿时间： | 2006-08-09 |
The Algorithm of Handwritten Chinese Character Recognition Based on Multiple HMM |
| |
| LIU Jian,LI Huifang,NIU Xinwei. The Algorithm of Handwritten Chinese Character Recognition Based on Multiple HMM[J]. China Information Security, 2007, 0(2): 75-77 |
| |
| Authors: | LIU Jian LI Huifang NIU Xinwei |
| |
| Abstract: | A new approach for handwritten Chinese character recognition based on multiple hidden Markov model classifiers and sub-region projection transform is proposed in the thesis. The object oriented program framework is con-structed by 4 HMMs for each Chinese character, and synthesizes the results of the 4 HMMs by equal proportion method, then gets the final recognition result. Experimental results show the efficiency of the proposed method. |
| |
| Keywords: | Chinese character recognition Feature extraction Multiple Hidden Markov Model Contour chain code |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
