| 基于遗传小波神经网络模型的脱机签名鉴定 |
| |
| 引用本文： | 侯海燕,张晓兰,郭静玉,段俊杰.基于遗传小波神经网络模型的脱机签名鉴定[J].河南科技大学学报(自然科学版),2009,30(3). |
| |
| 作者姓名： | 侯海燕 张晓兰 郭静玉 段俊杰 |
| |
| 作者单位： | 1. 河南科技大学,医学技术与工程学院,河南,洛阳,471003
2. 大唐洛阳热电厂,设备管理处,河南,洛阳,471000 |
| |
| 基金项目： | 河南省教育厅自然科学研究项目 |
| |
| 摘 要： | 根据脱机中文签名的特点,提出了一种用基于遗传算法的小波神经网络进行脱机中文签名鉴定的方法.基于遗传算法的小波神经网络模型结合了遗传算法的全局搜索能力和小波神经网络快速高精度的特点,克服了梯度下降法易陷入局部极小和引起振荡效应等缺点.该模型用遗传算法对小波神经网络的各参数进行前期优化训练,再用梯度下降法对参数进行二次优化.实验结果表明,将此模型用于脱机签名鉴定系统,与单独采用BP网络或小波神经网络相比,性能有明显的提高.
|
| 关 键 词： | 脱机签名鉴定 遗传算法 小波神经网络 |
Off-line Chinese Signature Verification Based on GA-WNN |
| |
| HOU Hai-Yan,ZHANG Xiao-Lan,GUO Jing-Yu,DUAN Jun-Jie.Off-line Chinese Signature Verification Based on GA-WNN[J].Journal of Henan University of Science & Technology:Natural Science,2009,30(3). |
| |
| Authors: | HOU Hai-Yan ZHANG Xiao-Lan GUO Jing-Yu DUAN Jun-Jie |
| |
| Institution: | 1.College of Medical Technology & Engineering;Henan University of Science & Technology;Luoyang 471003;China;2.Equipment Management;Datang Luoyang Thermal Power Plant;China |
| |
| Abstract: | Combining the advantage of genetic algorithm(GA) and wavelet neural network(WNN),a model of wavelet neural network based on genetic algorithm(GA-WNN) is proposed.In this model,GA is used to locate the initial parameters for WNN firstly,and then the gradient descent algorithm is employed to optimize these parameters.Used for the off-line Chinese signature verification,the experimental results prove that the GA-WNN model is more effective compared with the BP neural network or wavelet neural network respectiv... |
| |
| Keywords: | Off-line signature verification Genetic algorithm Wavelet neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
