| 基于AdaBoost的车牌字符快速识别方法研究 |
| |
| 引用本文： | 黄承清,高敬阳.基于AdaBoost的车牌字符快速识别方法研究[J].计算机与现代化,2010(9):140-143. |
| |
| 作者姓名： | 黄承清 高敬阳 |
| |
| 作者单位： | 北京化工大学信息科学与技术学院,北京,100029 |
| |
| 摘 要： | 采用动态自适应权重裁剪方法快速训练AdaBoost集成网络,通过在迭代过程中动态地寻找最优裁剪系数,缩短网络的训练时间,减少其个体数目。实验表明,该方法能够快速构建神经网络集成分类器,并取得良好的识别效果。
|
| 关 键 词： | 神经网络集成 字符识别 权重裁剪 自适应 快速训练 |
Fast License Plate Character Recognition Based on AdaBoost |
| |
| HUANG Cheng-qing,GAO Jing-yang.Fast License Plate Character Recognition Based on AdaBoost[J].Computer and Modernization,2010(9):140-143. |
| |
| Authors: | HUANG Cheng-qing GAO Jing-yang |
| |
| Affiliation: | (School of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China) |
| |
| Abstract: | Using dynamic adaptive weight trimming AdaBoost can train neural network ensemble faster,while searching the optimal cutting coefficient.It can reduce the training time and component network of the neural network ensemble system.Experiments show that the method can quickly build a neural ensemble classifier and achieve a good recognition effect. |
| |
| Keywords: | neural network ensemble character recognition weight trimming adaptive faster training |
| 本文献已被 维普 万方数据 等数据库收录！ |
