摘要
为了在各种智能交通管理系统中对车型进行准确的识别,通过对高速公路上以地感线圈为传感器而采集到的大量车辆样本进行数据分析和特征提取,提出以数据中的某些特征值作为识别依据,用隶属度函数充当知识库的识别方法,建立了车型识别专家系统, 通过实验获得了较满意的识别率。
For the purpose of recognizing vehicle pattern exactly, this article advances a method of designing the vehicle--recognizing expert system, in the basis of analyzing and feature extracting of a large amount of sampling data collected by loop sensors in expressway. This method relies on same characteristic value of the data, and considers the membership function to be its recognizing principles. And this system achieved a satisfying recognizing rate in experiments.
出处
《计算机测量与控制》
CSCD
2006年第4期472-473,487,共3页
Computer Measurement &Control
关键词
车型识别
专家系统
知识库
地感线圈
隶属度函数
vehicle recognition
expert system
knowledge database
loop sensor
membership function