| 基于支持向量机的车牌字符识别 |
| |
| 引用本文： | 陈来荣 冀荣华 徐宇. 基于支持向量机的车牌字符识别[J]. 公路交通科技, 2006, 23(5): 126-129 |
| |
| 作者姓名： | 陈来荣 冀荣华 徐宇 |
| |
| 作者单位： | 北京工业大学交通研究中心，中国农业大学教育部现代精细农业系统集成研究重点实验室，北华大学产业处 北京100022，北京林业大学工学院，北京100083，北京100083，吉林吉林132021 |
| |
| 基金项目： | 北京市科委科技项目(H020620220031) |
| |
| 摘 要： | 支持向量机(Support Vector Machines,简称SVM)能够有效地解决小样本学习、非线性及高维模式识别等问题。对此提出了在无特征提取情况下基于SVM的车牌字符识别方法,通过实验选定二次多项式作为核函数,并将基于SVM的车牌字符识别与基于BP神经网络的车牌字符识别进行了实验对比。结果表明,在训练样本较少的情况下,该系统具有较高的识别率和识别速度,并具有很好的分类推广能力。
|
| 关 键 词： | 模式识别 车牌字符识别 支持向量机 核函数 |
| 收稿时间： | 2005-01-27 |
Research on SVM-based License Plate Recognition |
| |
| CHEN Lai-rong, Jl Rong-hua, XU Yu. Research on SVM-based License Plate Recognition[J]. Journal of Highway and Transportation Research and Development, 2006, 23(5): 126-129 |
| |
| Authors: | CHEN Lai-rong Jl Rong-hua XU Yu |
| |
| Affiliation: | 1.TransportationResearch Center, Beijing University of Technology, Beijing 100022, China; 2.School of Technology, Beijing Forestry University, Beijing 100083, China; 3.Key Laboratory for Modem Precision Agriculture Integration Ministry of FAucation, China Agricultural University, Beijing 100083, China; 4.Industry Department, Beihua University, Jilin Jilin 132021, China |
| |
| Abstract: | Support vector machine(SVM) can be effective in small sample learning,non-linear and high dimensional recognition.A SVM-based recognition method is proposed for license plate identification without extracting features of the characters,and the quadratic polynomial is selected as kernel function by experiments.Finally,this method is compared with the method based on BP neuro-network by experiments.The results show that,in the case of small samples,the method works better than BP neuro-network in terms of recognition rate and speed. |
| |
| Keywords: | Pattern recognition License plate character recognition SVM Kernel function |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
