車(chē)輛牌照識(shí)別系統(tǒng)的研究－車(chē)牌字符識(shí)別子系統(tǒng)的設(shè)計(jì)與應(yīng)用

全文45頁(yè) 約21000字 論述翔實(shí)


摘 要
隨著我國(guó)經(jīng)濟(jì)的發(fā)展，汽車(chē)擁有量的急劇增加，道路交通成為我國(guó)重要的交通運(yùn)輸途徑，是國(guó)家大力發(fā)展的基礎(chǔ)設(shè)施，日益擁堵的城市交通需要用更先進(jìn)、更有效的交通管理、控制。利用電子信息技術(shù)來(lái)提高管理效率、交通效率和道路交通安全性的智能交通系統(tǒng)己成為當(dāng)前交通管理發(fā)展的主要方向。
車(chē)牌識(shí)別（License Plate Recognition，簡(jiǎn)稱(chēng)LPR）是智能交通中的關(guān)鍵技術(shù)之一?？梢詫?duì)車(chē)輛進(jìn)行自動(dòng)登記、驗(yàn)證、監(jiān)視、報(bào)警，進(jìn)而可以應(yīng)用在多種場(chǎng)合。
本論文利用神經(jīng)網(wǎng)絡(luò)技術(shù)，對(duì)車(chē)牌自動(dòng)識(shí)別系統(tǒng)進(jìn)行研究。在所拍攝的車(chē)牌照片中，先獲取車(chē)牌位置，再把車(chē)牌的各個(gè)字符分割，提取出各個(gè)字符的特征，再利用BP神經(jīng)網(wǎng)絡(luò)進(jìn)行識(shí)別，從而獲取車(chē)牌號(hào)碼。
本文主要對(duì)基于BP神經(jīng)網(wǎng)絡(luò)的車(chē)輛牌照識(shí)別系統(tǒng)進(jìn)行了較深入的研究和分析。首先介紹了汽車(chē)牌照識(shí)別系統(tǒng)的現(xiàn)狀；接下來(lái)對(duì)車(chē)牌識(shí)別系統(tǒng)的硬件和軟件設(shè)計(jì)進(jìn)行了介紹；其次對(duì)神經(jīng)網(wǎng)絡(luò)的構(gòu)成以及相關(guān)的理論進(jìn)行了討論，著重分析了神經(jīng)網(wǎng)絡(luò)的理論原理和在字符識(shí)別中的應(yīng)用方法；最后提出了車(chē)牌字符識(shí)別系統(tǒng)軟件設(shè)計(jì)的實(shí)現(xiàn)方法并給出了實(shí)驗(yàn)結(jié)果。

關(guān)鍵詞：車(chē)牌識(shí)別，字符識(shí)別，BP神經(jīng)網(wǎng)絡(luò)，特征提取

Research of License Plate Recognition System
Design and Application of License Plate Character Recognition Sub System
Abstract
With the fantastic spur in economy and rapid development of automobile, crowed urban traffic needs more advanced and more effective traffic administration and control system. So Intelligent Transportation Systems which makes use of electronic information technology becomes main direction of traffic administration.
License Plate Recognition is one of the critical techniques for the intelligent transportation systems. The system can automatically register, verify, monitor vehicle or report to the police by using automatic recognition of vehicle license plates.
The paper introduces a system in which neural networks has been applied to recognize the vehicle license plate. In the photo of vehicle where the license plates is involved, after making sure of the location of vehicle license plates, characters in the vehicle license plates are divided. Then, features of all characters have been drawn out and BP neural networks have been used to recognize all the characters, so the numbers of vehicle license plates are captured.
The paper gives a full introduction of the present situation of technologies in LPR system. And there are some introduces of software design and hardware design of an automobile LPR system. Then we analyze the structure of the neural network and emphatically discuss the theory of it. At last we introduce the theoretical foundation and method of the software design of the LPR system in detail.

Key words: license plate recognition, characters recognition, BP neural network, features extraction

目 錄

1 緒論 6
1.1智能交通系統(tǒng) 6
1.2 汽車(chē)牌照自動(dòng)識(shí)別系統(tǒng) 7
1.3 車(chē)牌識(shí)別技術(shù)研究現(xiàn)狀 9
1.4 論文主要內(nèi)容 10
2 車(chē)輛牌照識(shí)別系統(tǒng)的結(jié)構(gòu) 11
2.1 車(chē)牌識(shí)別系統(tǒng)的總體結(jié)構(gòu) 11
2.2系統(tǒng)的硬件構(gòu)成及其原理 12
2.2.1 全天候圖像采集技術(shù) 12
2.3.2 數(shù)字信號(hào)處理器（DSP） 14
2.3系統(tǒng)的軟件構(gòu)成及其原理 14
2.3.1車(chē)牌預(yù)處理 15
2.3.2牌照的定位與分割 16
2.3.3字符的分割 17
2.3.4字符的識(shí)別 19
3 神經(jīng)網(wǎng)絡(luò)的原理概述 21
3.1神經(jīng)網(wǎng)絡(luò)理論概述 21
3.2 使用BP神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)及其學(xué)習(xí)算法 22
3.2.1 多層網(wǎng)絡(luò)結(jié)構(gòu) 23
3.2.2 BP神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)公式推導(dǎo) 24
3.2.2 BP神經(jīng)網(wǎng)絡(luò)的訓(xùn)練 30
3.2.3 BP神經(jīng)網(wǎng)絡(luò)的設(shè)計(jì)分析 30
4 汽車(chē)牌照的識(shí)別方法 32
4.1 車(chē)牌定位分割 32
4.2 字符分割 33
4.3 基于神經(jīng)網(wǎng)絡(luò)的字符識(shí)別方法 35
4.3.1 字符識(shí)別的預(yù)處理 35
4.3.2 特征編碼 37
4.3.3 BP神經(jīng)網(wǎng)絡(luò)的構(gòu)建 39
5 車(chē)牌字符識(shí)別系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn) 40
5.1 創(chuàng)建一個(gè)BP神經(jīng)網(wǎng)絡(luò) 40
5.2 設(shè)定訓(xùn)練參數(shù) 40
5.3 無(wú)噪聲情況下訓(xùn)練網(wǎng)絡(luò) 41
5.4 含噪聲情況下訓(xùn)練網(wǎng)絡(luò) 42
5.5 字符預(yù)處理 43
5.6 特征編碼 43
5.7利用神經(jīng)網(wǎng)絡(luò)識(shí)別字符 44
結(jié) 論 45
致 謝 46
參考文獻(xiàn) 47

部分參考文獻(xiàn)

[7] 聞新、周露等.Matlab神經(jīng)網(wǎng)絡(luò)應(yīng)用設(shè)計(jì).科學(xué)出版社，2000.207～212
[8] 胡守仁.神經(jīng)網(wǎng)絡(luò)應(yīng)用技術(shù).國(guó)防科技大學(xué)出版社.1999.1～217
[9] 施鴻寶.神經(jīng)網(wǎng)絡(luò)及其應(yīng)用.西安交通大學(xué)出版社.1993.1～44
[10] Lotufo R A, Morgan A D, Johnson A S. Automatic number-plate recognition. Image Analysis for Transport Application, IEE Colloquium ,1990,vol.6,1～6