机构地区: 广东技术师范学院
出 处: 《公路交通科技》 2006年第7期125-129,共5页
摘 要: 针对当前车型分类系统识别率不高的问题,提出了用4条共聚物压电轴传感器组成的传感器阵列获取车型特征参数的方法。该方法利用传感器的特殊空间分布与车辆各车轴到达各传感器时刻的时空相关性,对采集到的信息进行数据融合,获得了多达11种车型特征参数。分类识别中用模糊统计的方法对大量的样本车型特征参数处理后,构造了隶属函数,最后采用了D-S证据理论对车型进行分类。实践证明该方法具有较高的识别率。 Aimed at solving the problem of less effective vehicle classification system, this paper proposes a method that utilizes 4 copolymer-piezoelectric-axis-sensors to make up the sensor array to get the vehicle type. This method uses the space dependencies of sensors" special distribution and the timing of vehicle axis contacting the corresponding sensor, integrates information collected into data and gets up to 11 vehicle type feather parameters. Fuzzy statistical algorithm is adopted to construct membership function after handling a great quantity of vehicle type data, and finally vehicle type is classified by using D-S method .The practice demonstrates that method has better recognition rate.