摘 要

在不停车收费系统(ETC, Electronic Toll Collection)引入中国之前，为了能够逃避交警的追查，部分车辆会贴假牌照、污损拍照甚至不贴车牌上路。套牌车问题已经给人们的生活带来了巨大的安全隐患。而后来，随着ETC的引入和普及，为了减少过桥过路的收费，出现了不少不法分子把其他不同车型车的车载单元(OBU, On Board Unit)安装在自己的车上的现象，套牌车问题更呈现出愈演愈烈的趋势。所以，为了国家的税收，为了维护交通秩序，为了减少人民生活的安全隐患，研究出套牌车识别的软硬件系统已经成为迫在眉睫的任务。

本文对ETC系统的现状和套牌车问题进行深入分析，提出一种基于数据库的套牌车识别系统管理软件的设计方法，旨在与硬件系统配合，解决套牌车违法问题。本文首先对ETC系统的国内外发展现状进行分析，得出套牌车违法的类型及识别常用的手段。比如，套牌车识别大部分采用图像处理得到车辆信息并与车载单元中存入的信息作对比，可以从车牌号、车型、车辆颜色、车主信息等方面着手。若其中一个或多个方面不一致，则可以判定这辆车是套牌车。然后通过分析主流数据库MySQL、SQL Server等的特点，结合ETC系统是大型系统这个特征，最终决定用中大型数据库SQL Server作后台数据库。用户界面则使用Java语言在Eclipse平台上编程实现，达到友好、严谨、实用的基本要求。在实现基本功能之后，对软件仔细测试并加以改进，达到稳定可靠运行要求。测试结果表明，套牌车识别系统管理软件可以很好地与硬件结合，通过比较图像处理得到的信息以及车载单元中存入的信息得出正确的判别结果，并且能在事后根据一定条件查询符合条件的比较信息详情，因此本设计在改善套牌车问题方面具有实用性。

关键词：套牌车，SQL Server，Java，图像处理

Abstract

Before ETC system (Electronic Toll Collection) was introduced to China, in order to avoid the tracing of the traffic police, some vehicles choose to drive on the road with fake license, incomplete license or even without any license plate. Deck car problem has become a potential great security problem to people’s lives. Later, with the introduction and popularization of ETC, in order to reduce the crossing of bridges’ and roads’ fees, there has been a trend that many criminals prefer to drive on road with other cars’ OBUs( On Board Unit) instead of their owns. Such problem has become more and more intensive. Therefore, in order to protect the national tax, in order to maintain traffic order, in order to reduce security risks in life, the development of deck card identification software and hardware system has become an urgent task.

This paper analyzes the status of the ETC system and the problem of the deck car, and puts forward a design method of the database-based deck identification system management software, which is designed to cooperate with the hardware system to solve the problem of deck defense. This paper firstly analyzes the present situation of ETC system at home and abroad, and figures out the types of illegal vehicles and the common means of identification. For example, the identification of deck cars depends mostly on the comparison of the vehicle information from image processing and the information stored in OBU, from the license plate number, model, vehicle color, owner information and other aspects. If one or more of them is inconsistent, you can determine that the car is a deck car. And then through the analysis of characteristics of the mainstream database MySQL, SQL Server and other databases, combined with the feature that ETC system is a large system, the paper finally decide to use large database SQL Server as a back-end database. User interface is achieved in the Eclipse, using Java language, to live up to the basic requirements of friendship, preciseness, practicability. After the realization of the basic functions, the software carefully tested and improved to achieve stable and reliable operation requirements. After the realization of the basic functions, the software was carefully tested and improved to achieve stable and reliable operation requirements. The test results show that the deck card identification system management software can be well combined with the hardware, by comparing the information obtained by the image processing and the information stored in the vehicle unit to get the correct judgment result. After the event, in accordance with certain conditions, the related staff can search for the detailed information about the comparison. So the design in this paper is practical in the improvement of deck car problem.

Keywords：deck car，SQL Server，Java，image process

目 录

第1章 绪论 1

1.1课题的研究背景和意义 1

1.2 国内外研究的现状 1

1.3 论文主要内容 2

第2章 套牌车识别系统整体方案设计 3

2.1 需求分析 3

2.2 可行性分析 5

第3章 套牌车识别软件数据库设计 6

3.1 数据字典设计 6

3.2 关系模型设计 8

3.2.1 概念结构设计 8

3.2.2 概念模型向关系模型转换 9

3.2.3 关系模型服从的范式 10

3.3 根据关系模型构建数据库 11

3.4 本章小结 14

第4章 套牌车识别软件判别套牌车的流程图 15

4.1 总体设计 15

4.2主界面设计 16

4.3 工作界面设计 18

4.4 管理界面设计 22

4.5 ETCManagement数据库与用户界面连接 26

第5章 套牌车识别系统软件测试 28

5.1 套牌车识别功能测试 28

5.2 套牌车查询功能测试 29

第6章 结论 32

参考文献 31

致谢 32

第1章 绪论

1.1课题的研究背景和意义

随着汽车工业蓬勃发展，人均拥有汽车量呈指数式上涨。因此，汽车在人们的日常生活中和工业上的交通运输中均发挥着越来越重要的作用。随之而来是以套牌车为主的违法犯罪活动，这除了立法缺位方面问题，在管理方面也存在着严重不足。目前在国内套牌车的识别主要是依靠警察判别或者民众举报，极大地浪费了警力也没有很好的打击力度^[1]。随着不停车收费技术（ETC）在中国的推广，车辆不仅可以由车牌标识，更可以由贴在车辆前端的电子标签标识，套牌车的相关违法犯罪行动也变得越发复杂。目前，使用伪造、变造的机动车号牌、使用其他车辆的机动车号牌、贴上其他车辆OBU的机动车、使用污损、故意遮挡的机动车号牌的机动车都可以归于套牌车这一类，而“套牌”也从普通的套用民用车牌发展到了套用警牌、军牌的地步^[2]。因此，研发出可以识别、管理套牌车的软硬件系统更是到了迫在眉睫的地步。

本文的目的就在于编程实现套牌车识别系统管理软件，使其可以与硬件配合，可在车辆通过ETC通道时通过比较自动判断出该车是否为套牌车并记录判断依据,从管理层的角度利用编程技术改善这一违法现象。