【摘要】： 隨著我國(guó)城市化進(jìn)程加速及汽車(chē)行業(yè)的迅猛發(fā)展,為加強(qiáng)道路車(chē)輛管理、提高運(yùn)輸效率,增強(qiáng)交通安全,車(chē)輛牌照自動(dòng)識(shí)別技術(shù)(License Plate Recognition,LPR)作為智能交通系統(tǒng)的重要組成部分,應(yīng)用前景廣泛。以車(chē)牌識(shí)別為基礎(chǔ),獲取車(chē)輛信息后在多種場(chǎng)合都可有效使用,如高速公路收費(fèi)系統(tǒng),道路監(jiān)控系統(tǒng),小區(qū)、車(chē)場(chǎng)收費(fèi)管理系統(tǒng),交通流量統(tǒng)計(jì)等,以進(jìn)行自動(dòng)登記、檢測(cè)、驗(yàn)證、報(bào)警等。 理論上,車(chē)牌識(shí)別系統(tǒng)包括車(chē)牌圖像預(yù)處理、車(chē)牌定位、車(chē)牌字符切割和字符識(shí)別四個(gè)主要部分,其中車(chē)牌定位是后續(xù)工作的基本保證。本文在研究分析近年來(lái)一些典型的車(chē)牌定位算法的基礎(chǔ)上,提出針對(duì)國(guó)內(nèi)任何類(lèi)型機(jī)動(dòng)車(chē)輛牌照,在復(fù)雜光照條件下都可識(shí)別的算法。本論文研究的主要內(nèi)容和創(chuàng)新點(diǎn)包括: 在車(chē)牌圖像定位方面,采用基于灰度圖像形態(tài)學(xué)處理的方法,利用頂帽變換進(jìn)行車(chē)牌定位的算法,并構(gòu)造適宜的結(jié)構(gòu)元素。實(shí)際應(yīng)用中,車(chē)體由于廣告標(biāo)識(shí)、光線(xiàn)干擾等情況產(chǎn)生的偽車(chē)牌區(qū)域。而后為加強(qiáng)定位精度,提出對(duì)車(chē)牌區(qū)域利用線(xiàn)性投影技術(shù)的二次定位,大大提高了車(chē)牌位置的精確度。 在車(chē)牌定位軟件系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)上,采用在VC++6.0環(huán)境下使用Open CV進(jìn)行開(kāi)發(fā)的方法,系統(tǒng)實(shí)現(xiàn)的功能則包括車(chē)輛圖像的導(dǎo)入與編輯、車(chē)輛圖像二值化及預(yù)處理、車(chē)牌定位等。
[Abstract]:With the acceleration of urbanization and the rapid development of automobile industry in China, in order to strengthen the management of road vehicles, improve the transport efficiency, enhance the traffic safety, vehicle license plate automatic identification technology (License Plate Recognition, As an important part of intelligent transportation system, LPR (Intelligent Transportation system) has a wide application prospect. Based on license plate recognition, vehicle information can be used effectively on many occasions after obtaining vehicle information, such as highway toll system, road monitoring system, district, parking lot toll management system, traffic flow statistics and so on, for automatic registration and detection. Verification, alarm, etc. Theoretically, the license plate recognition system consists of four main parts: license plate image preprocessing, license plate location, license plate character cutting and character recognition, in which license plate location is the basic guarantee of the follow-up work. Based on the research and analysis of some typical license plate location algorithms in recent years, this paper proposes an algorithm which can be recognized under complex illumination conditions for any type of license plate of domestic motor vehicles. The main contents and innovations of this paper include: in license plate image location, the method based on gray image morphology processing is adopted, and the top cap transformation is used to carry out license plate location algorithm, and the appropriate structural elements are constructed. In practical application, the vehicle body due to advertising signs, light interference and other conditions of the false license plate area. Then, in order to enhance the positioning accuracy, the linear projection technique is used to locate the license plate area, which greatly improves the accuracy of the license plate location. In the design and implementation of license plate positioning software system, the method of using Open CV in VC 6.0.The functions of the system include the import and edit of vehicle image, the binarization of vehicle image and the pretreatment of vehicle image, and the functions of the system include import and edit of vehicle image, binary of vehicle image and pretreatment of vehicle image. License plate location etc.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2010
【分類(lèi)號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 李小平,任江興,楊德剛;車(chē)牌識(shí)別系統(tǒng)中若干問(wèn)題的探討[J];北京理工大學(xué)學(xué)報(bào);2001年01期

2 袁寶民,于萬(wàn)波,魏小鵬;汽車(chē)牌照定位研究綜述[J];大連大學(xué)學(xué)報(bào);2002年02期

3 路小波;張光華;;基于二值圖像的車(chē)牌精確定位方法[J];東南大學(xué)學(xué)報(bào)(自然科學(xué)版);2005年06期

4 白洪亮,婁正良,鄒明福,劉昌平;復(fù)雜背景下基于形態(tài)學(xué)的車(chē)牌識(shí)別系統(tǒng)[J];公路交通科技;2004年10期

5 金玲玲,廖芹,汪劉一;汽車(chē)牌照的提取方法[J];華南理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2002年07期

6 郭捷,施鵬飛;基于顏色和紋理分析的車(chē)牌定位方法[J];中國(guó)圖象圖形學(xué)報(bào);2002年05期

，

本文編號(hào)：2459438