毕业设计（论文）-车牌识别系统研究.doc

摘要
随着现代交通运输的迅猛发展,以及汽车的大量普及,车牌识别技术的应用已经成为交通管理部门重要的管理手段。本文主要研究了车牌字符的识别技术,并将BP神经网络应用于车牌的识别,在车牌图像进行预处理的基础上,重点讨论了用BP神经网络方法对车牌字符的识别。BP网络包含了神经网络理论中最精华的部分,由于其结构简单、可塑性强,得到了广泛的应用,特别它的数学意义明确、步骤分明的学习算法更使其具有更广泛的应用背景。本文介绍的方法是建立在首先将训练样本做图像预处理,再对车牌上的字符进行分割,得到单个字符的基础上,对大小不一的字符做归一化后,得到二值图像,作为输入矢量。把这个特征向量送到BP网络中进行训练,得到了训练好的权值。然后输入要识别的图片,进行识别。程序结果显示该方法具有一定的可靠性和可应用性。
关键字:LPR;字符识别; BP神经网络;
ABSTRACT
With the rapid development of the modern transport and the large mumber of popular car,license plate recognition technology have e an inportant management tool of the traffic the discernment to the number plate character, this text applies BP work to the automatic discernment of the automobile number plate, on the basis that the number plate picture goes on in advance treated , is it use BP work method to car discernment , license plate of character to discuss includes word theory in essence,because of its simple structure ,strong plastic,it is widely its mathematical meaning clear,distinct step in the learning algorithm is more to the wider use of background. Will train samples to do the pretreatment of the picture at first, character in number plate cut apart, get the individual character. Then we get a receive characteristic vector quantity like this to put, what the vector quantity of this characteristic is written down is the characteristic of the the characteristic vector quantity work train, get good right value of trainin. Open picture discerned to want, go on to picture in advance treated to can discern. It shows that the validity of this method .The results of the procedures showed that the method is reliability.
Key words:The number plate discerning;The character discerning;LPR;BP work;
目录
第一章概述 1
基本概念 1
车牌字符识别简介 4
车牌字符识别发展概况 4
车牌字符识别系统用到的方法 4
国内外研究现状和发展趋势 7
第二章 神经网络简介 9
9
9
几种常用的作用函数 10
10
基本模型 10
前向网络 11
Hebb学习规则 11
2