杰出人士
- 积分
- 15998
- 威望
- 397
- 贡献
- 15525
- 兑换币
- 2
- 注册时间
- 2010-3-14
- 在线时间
- 38 小时
|
第26 卷 第2 期
2009 年2 月
公 路 交 通 科 技
Journal of Highway and Transportation Research and Development
Vol126 No12
Feb12009
文章编号: 1002O0268 (2009) 02O0108O06
收稿日期: 2008O02O16
基金项目: 国家自然科学基金资助项目(50578064) , 广州市科技攻关重点项目(2007Z2 - D3111)
作者简介: 林培群(1980 - ) , 男, 博士, 讲师, 研究方向为智能交通、智能信息处理1 (lpqemail @1631com)
感应线圈信号曲线的处理方法
及其在车型识别中的应用
林培群, 徐建闽
(华南理工大学 土木与交通学院, 广东 广州 510640)
摘要: 为提高车型识别的准确率, 提出信号处理的方法: 首先采用快速小波变换抽取原始曲线的近似轮廓, 从而削弱
了噪声等细节信息的干扰, 并实现2 n 倍的数据压缩; 其次利用五点三次滤波器对上一步获得的近似曲线进行平滑,
使得离散信号更接近于真实的连续信号; 在此基础上提出基于峰值时间差的强鲁棒性车长检测方法; 最后, 根据形态
匹配的要求, 采用Akima 插值法进行信号尺度的归一化处理。车型识别试验采用观测点的实测数据, 并将结果与其他
文献进行比较, 说明所提出的方法的有效性。
关键词: 智能运输系统; 车型识别; 感应线圈; 信号处理; 特征提取
中图分类号: U491 文献标识码: A
Proce ssing Methods of Signal Trace in Inductive Loop Detector and It s
Application in Vehicle Recognition
LIN Peiqun , XU Jianmin
(School of Civil Engineering and Transportation , South China University of Technology , Guangzhou Guangdong 510640 , China)
Abstract : Some methods of signal processing were proposed for improving the precision of vehicle recognition1First ,
fast wavelet transform (FWT) which realizes 2n compression ratio was employed to extract the approximate profile of
original traces to reduce the molesting of detailed information , such as noises1Second , 5OpointO3O time filter which
makes the discrete signal series close to the real continuous one was used to smooth the approximate trace obtained
by FWT1Furthermore , robust vehicleOlength detection which adopts the time point difference of traceOpeaks was
studied and presented1Finally , Akima method was employed to uniform the signal scaling for trace configuration
matching1Real data collected from traffic observation station were used in the vehicle recognition experiment1The
experimental results were compared with those of other literatures and demonstrated that the proposed methods are
efficient1
Key words : Intelligent Transport Systems ; vehicle recognition ; inductive loop detector ; signal processing ; fea2
ture extraction
0 引言
感应线圈作为应用最广泛的交通检测器, 在全天
候工作稳定性方面有着视频等检测系统无法比拟的优
势。根据LC 振荡电路的原理, 机动车经过感应线圈
时会产生相应的信号曲线, 不同车型由于底盘结构、
高度等因素的差异所产生的曲线并不相同, 这就为车
型分类提供了依据。特别的, 区分轮廓相似的客车和
货车对以外形尺寸为分类依据的大部分检测技术来说
是个难以逾越的障碍, 而大量试验已经证明, 感应线 |
|