欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对现有车辆分类算法精度不高的问题，基于既有的车辆检测算法，提出一种基于决策树初次分类，K邻近算法细分的车辆分型算法，实验证明算法行之有效并取得了较好的分类效果．为了验证算法的可行性，还设计实现了停车管理系统（Parking-Management System，PMS），包括终端、路由、基站节点的硬件、软件，协议栈，网络的架构以及上位机的整套监控软件．对PMS运行数月得到的数据进行分类，结果显示，分类准确率显著提高．
英文摘要：
In order to solve the low accuracy of the current vehicle classifying algorithms, this paper presents a vehicle classification algorithm based on the existed vehicle detecting algorithm, which classifies the vehicles roughly by a decision tree firstly and then decides the types of the vehicles exactly using the KNN algorithm. Proved by some experiments,this algorithm works well and has a good classifying performance. Moreover,in order to test and verify the feasibility of this algorithm,this paper sets up a PMS（Parking Management System）which includes the design and implementation of the hardware and software of end devices, routes, base station, the ZigBee protocol stack, the network structure and the whole monitoring software of the server. By analyzing and classifying the PMS data, which are collected from the system run for several months, the proposed algorithm gets higher vehicle classifying accuracy than the existing methods.
同期刊论文项目
同项目期刊论文
期刊信息
位置：