| 基于神经网络方法的手写体数字识别 |
| |
| 引用本文： | 高鹏,谭红,杜华,赵大威. 基于神经网络方法的手写体数字识别[J]. 吉林大学学报(工学版), 1997, 0(1) |
| |
| 作者姓名： | 高鹏 谭红 杜华 赵大威 |
| |
| 作者单位： | 哈尔滨工业大学，吉林工业大学 |
| |
| 摘 要： | 利用神经网络技术对无限制手写体数字进行识别。在选取合适的特征基础上，对图像进行分域处理，每一小域对应输入层的一个神经元。测试结果表明，该识别方案具有很强的抗畸变、抗旋转能力。
|
| 关 键 词： | 神经网络 特征 识别 |
A Method Based on Neural Network in Recognition of Handwritten Numerals |
| |
| Gao Peng. A Method Based on Neural Network in Recognition of Handwritten Numerals[J]. Journal of Jilin University:Eng and Technol Ed, 1997, 0(1) |
| |
| Authors: | Gao Peng |
| |
| Affiliation: | Gao Peng (Harbin Institute of Technology) Tan Hong Du Hua Zhao Dawei (Jilin University of Technology) |
| |
| Abstract: | This paper introduces neural network to recognize unconstrained handwritten numbers. On the basis of appropriate feature extraction, image segmentation is made. Every part corresponds to a neural cell of input layer. Experimental results show that the method has the ability of rotational variance resistance. |
| |
| Keywords: | neural network feature recognition |
| 本文献已被 CNKI 等数据库收录！ |
