| 基于自编码神经网络重构的车牌数字识别 |
| |
| 引用本文： | 刘高平.基于自编码神经网络重构的车牌数字识别[J].光电子．激光,2011(1):144-148. |
| |
| 作者姓名： | 刘高平 |
| |
| 作者单位： | 浙江万里学院电子工程研究所;西安交通大学电子与信息工程学院; |
| |
| 基金项目： | 国家自科学基金资助项目(60641002); 工业控制技术国家重点实验室开放课题基金资助(ICT1004) |
| |
| 摘 要： | 提出了一种基于自编码神经网络重构的车牌数字识别方案.首先对车牌图像进行预处理,利用车牌字符的原图和Gabor特征作为自编码神经网络的输入进行识别实验.然后对每个车牌字符构造一个自编码神经网络,利用训练样本进行图像的重构训练,并根据训练得到的网络权值重构出训练样本集中的各个字符图像或特征.最后,将测试样本输入到每个自编码...
|
| 关 键 词： | 自编码神经网络 车牌数字 字符识别 重构 |
Character recognition of license plate using autoencoder neural network reconstruction |
| |
| LIU Gao-ping.Character recognition of license plate using autoencoder neural network reconstruction[J].Journal of Optoelectronics·laser,2011(1):144-148. |
| |
| Authors: | LIU Gao-ping |
| |
| Affiliation: | Electronic Engineering Institute,Zhejiang Wanli University,Ningbo 315100,China; |
| |
| Abstract: | A character recognition scheme for the license plate based on several autoencoder reconstruction neural networks is proposed in this paper.The method needs some preprocessing steps at first,and original license plate character images and the Gabor features are taken as the input of the autoencoder neural network for recognition.In the recognition system,the neural networks that train the network weights for reconstruction of input are established for each character.The sample to be tested is input to each n... |
| |
| Keywords: | autoencoder neural network license plate number character recognition reconstruction |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《光电子．激光》浏览原始摘要信息 |
|
点击此处可从《光电子．激光》下载全文 |
