| “有效行”特征对手写体字符的识别 |
| |
| 引用本文： | 王贵新,刘建胜,居琰,汪同庆,彭健,杨波.“有效行”特征对手写体字符的识别[J].电子科技大学学报(社会科学版),2001(3). |
| |
| 作者姓名： | 王贵新 刘建胜 居琰 汪同庆 彭健 杨波 |
| |
| 作者单位： | 重庆后勤工程学院!重庆400014(王贵新)，重庆大学光电工程学院!重庆400000(刘建胜，居琰，汪同庆，彭健，杨波) |
| |
| 摘 要： | 从无限制的手写体数字的结构出发，提出“有效行”特征的概念及其提取算法。该特征具有维数小、平移不变、字符小角度旋转不变等特点。建立了相应的字符特征库，利用ＢＰ网络对样本字符进行研究，通过大量的手写体数字识别测试表明：该方法在识别速度上优于利用单一的矩特征、小波特征等传统识别方法；在误识率方面也优于一般的单一识别方法。
|
| 关 键 词： | 特征提取 识别 神经网络 手写体 |
Handwritten Numerals Recognization Based on "Effective-rows",Feature |
| |
| Wang Guixin.Handwritten Numerals Recognization Based on "Effective-rows",Feature[J].Journal of University of Electronic Science and Technology of China(Social Sciences Edition),2001(3). |
| |
| Authors: | Wang Guixin |
| |
| Institution: | Wang Guixin (Institute of Logistic and Engineering of Chongqing Chongqing 400014) Liu Jiansheng Ju Yan Wang Tongqing Peng Jian Yang Bo (Chongqing University Chongqing 400000) |
| |
| Abstract: | The concept and extracting method of the effective-rows feature, which possesses low dimension, shift-invariance and rotation-invariance under little angle is proposed on the basis of the structure of unconstraint and handwritten numerals in this paper. Its features is also constructed. According to the result of widely verifying recognition BP network, it is demonstrated that its recognitionspeed is superior to the traditional recognition methods only based on the feature of moment or wavelet and its recognition-extract is higher than some conventional methods. |
| |
| Keywords: | feature extraction recognition neural network handwritten |
| 本文献已被 CNKI 等数据库收录！ |
