| 基于小波变换和ART网络的手写数字识别 |
| |
| 引用本文： | 张捷,寇雪芹,封俊红. 基于小波变换和ART网络的手写数字识别[J]. 计算机测量与控制, 2004, 12(11): 1093-1095 |
| |
| 作者姓名： | 张捷 寇雪芹 封俊红 |
| |
| 作者单位： | 西安建筑科技大学,管理学院,陕西,西安,710055;西安建筑科技大学,机电学院,陕西,西安,710055 |
| |
| 基金项目： | 陕西省教育厅专项基金资助项目(01JK201) |
| |
| 摘 要： | 由于小波变换能有效地提取字符的结构特征，自适应共振(ART)网络有很好的学习能力。将二者结合起来，用小波变换抽取特征、用自适应共振ART网络作模式分类器来识别手写数字。实验证明该方法有很高的识别率，能够有效地进行手写数字的分类，可以满足实际应用。
|
| 关 键 词： | 手写数字识别 小波变换 ART网络 |
| 文章编号： | 1671-4598(2004)11-1093-03 |
| 修稿时间： | 2004-01-24 |
Handwritten Digit Recognition Based on Wavelet Transform and ART Neural Networks |
| |
| Zhang Jie,Kou Xueqin,Feng Junhong. Handwritten Digit Recognition Based on Wavelet Transform and ART Neural Networks[J]. Computer Measurement & Control, 2004, 12(11): 1093-1095 |
| |
| Authors: | Zhang Jie Kou Xueqin Feng Junhong |
| |
| Affiliation: | Zhang Jie~1,Kou Xueqin~2,Feng Junhong~1 |
| |
| Abstract: | Because of that wavelet transform can effectively extract the features of character construction and adaptive resonance theory (ART) neural networks has a nice learning ability, the two aspects are combined to recognize handwritten digit using wavelet transform to extract features and adaptive resonance theory (ART) neural networks for classification. The result of experiment shows high recognition rate, which indicates that the methods can effectively classify hand written digit and be put into practical use. |
| |
| Keywords: | handwritten digit recognition wavelet transform ART neural networks |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
