车牌识别系统中定位算法的研究
|
摘要
随着信息技术和智能技术的发展,交通管理系统的信息化、智能化已成为发展的趋势。车牌识别系统(LPR)是智能交通系统(ITS)的核心组成部分,在现代交通收费管理系统中发挥着举足轻重的作用。作为一个综合的实时计算机视觉系统,车牌识别技术主要包括车牌定位和车牌识别两个部分。而车牌的定位更是该系统的关键之一,由于图像场景的复杂性以及车牌位置和图像质量的不可预知性,车牌定位系统一直都未做到令人满意,所以车牌定位算法一直是该领域的研究热点。
本文采用基于数学形态学和连通区域分析的综合车牌定位方法,主要思想是将检测车牌的文本信息作为定位车牌的突破口,再结合车牌自身一些特征来进行精确定位。主要是通过以下几个步骤来实现:首先将车牌灰度图像按照四个方向的结构元素分别进行高帽变换和低帽变换处理,并从这四个方向的处理结果中计算出高帽变换和低帽变换的能量图;其次根据车牌文本中的各个字符在高度、大小上具有相似性,且具有连贯性这个特征,进行第一次连通区域分析,粗略地去除伪文本区域,然后利用Hough变换找出同一方向上的车牌文本区域的中心线,将其连通起来。最后将两张结果图相加起来,进行第二次连通区域分析,根据车牌固有特征来精确提取出车牌区域。
通过对采集于各种真实环境的图像进行实验,结果表明,本文所采用的方法能达到较好的车牌定位效果,具有一定的鲁棒性和实时性。
With the development of information technology and intelligence technology, the informatization and intelligentizing of traffic management is the trend. License Plate Recognition system (LPR) is the core of Intelligent Traffic System (ITS). It is very important in modern traffic management systems. As a comprehensive real-time computer vision system, the vehicles license plate recognition (LPR) mainly includes the division of license plate localization and recognition of character. Because of the complexity of the scene, the unpredictability of license position and picture quality, the license plate localization division system is still something unsatisfied. Therefore the license plate localization division algorithm is always a hot spot of research in this field.
A hybrid license plate location scheme that is based on mathematical morphology is presented in this paper, which is mainly implemented as following several steps. At first, we get the gray image of the license plate do the top-hat and the bottom-hat transformation, then, in first CCA (Connected component analysis) stage, we get rid of those non-text roughly according to the characters of license plate have the similar feature in height and size, and they also are connected with each other. Later, we connect these plate characters into one component based on finding the center line of them using Hough transform. Furthermore, the second CCA stage is accurately extracting the license plate according the real feature of the license plate.
Based on the experiments dealing with images taken under various real world conditions, the results prove that this proposed method can relatively locate license plate. It shows that the performance of the system is promising, robust and timely.
本文采用基于数学形态学和连通区域分析的综合车牌定位方法,主要思想是将检测车牌的文本信息作为定位车牌的突破口,再结合车牌自身一些特征来进行精确定位。主要是通过以下几个步骤来实现:首先将车牌灰度图像按照四个方向的结构元素分别进行高帽变换和低帽变换处理,并从这四个方向的处理结果中计算出高帽变换和低帽变换的能量图;其次根据车牌文本中的各个字符在高度、大小上具有相似性,且具有连贯性这个特征,进行第一次连通区域分析,粗略地去除伪文本区域,然后利用Hough变换找出同一方向上的车牌文本区域的中心线,将其连通起来。最后将两张结果图相加起来,进行第二次连通区域分析,根据车牌固有特征来精确提取出车牌区域。
通过对采集于各种真实环境的图像进行实验,结果表明,本文所采用的方法能达到较好的车牌定位效果,具有一定的鲁棒性和实时性。
With the development of information technology and intelligence technology, the informatization and intelligentizing of traffic management is the trend. License Plate Recognition system (LPR) is the core of Intelligent Traffic System (ITS). It is very important in modern traffic management systems. As a comprehensive real-time computer vision system, the vehicles license plate recognition (LPR) mainly includes the division of license plate localization and recognition of character. Because of the complexity of the scene, the unpredictability of license position and picture quality, the license plate localization division system is still something unsatisfied. Therefore the license plate localization division algorithm is always a hot spot of research in this field.
A hybrid license plate location scheme that is based on mathematical morphology is presented in this paper, which is mainly implemented as following several steps. At first, we get the gray image of the license plate do the top-hat and the bottom-hat transformation, then, in first CCA (Connected component analysis) stage, we get rid of those non-text roughly according to the characters of license plate have the similar feature in height and size, and they also are connected with each other. Later, we connect these plate characters into one component based on finding the center line of them using Hough transform. Furthermore, the second CCA stage is accurately extracting the license plate according the real feature of the license plate.
Based on the experiments dealing with images taken under various real world conditions, the results prove that this proposed method can relatively locate license plate. It shows that the performance of the system is promising, robust and timely.
引文
[1]王笑京,ITS在中国的发展,http://www.itsc.com.cn.
[2]蔡文沁,我国智能交通系统发展的战略构想.交通运输系统工程与信息,2003年2月,Vol 3, No.l:16-22.
[3] Javier Cano and Juan-Carlos P′erez-Cort′es F.J. Perales et al. (Eds.): Vehicle License Plate Segmentation in Natural Images, IbPRIA 2003, LNCS 2652, 2003._c Springer-Verlag Berlin Heidelberg 2003:142-149.
[4]尹令,车牌自动识别系统的研究及其在VC++中的实现,[硕士学位论文],湘潭,湘潭大学计算控制技术与控制理论应用,2003:1-3.
[5]陈旭梅,于雷,郭继孚等,美国智能交通系统ITS的近期发展综述,中外公路,2003 Vo1.23, No.2: 9-120.
[6] Barroso J,Bulas-Cruz J & Dagless E L,Real-Time Number Plate Reading[J]. 4th IFAC Workshop on Algorithms and Architectures for Real-Time Control,1997.
[7] R.Parisi. Car Plate Recognition by Neural Networks and Image Processing[J]. IEEE International Symposium on Industrial Electronics, 1998:605-610.
[8] C.H.Coetzee,et al. PC Based Number Plate Recognition Systems, In Proc.IEEE International Symposium on Industrial Electronics. 1999,1(1):605-610.
[9] C.Rittscher, J.Kato, S.Joga, and A.Blake., A Probabilistic Background Model for Tracking. 2000, 2(2): 63-67.
[10]刘效静,成瑜. ,汽车牌照自动识别技术研究,[J].南京航天航空大学学报, 1998,30(5):573-575.
[11]牛欣,沈兰荪,基于特征的车辆牌照定位算法,[J].交通与计算机,. 2001,1(81):31-33.
[12]刘智勇,刘迎建,车牌识别(LPR)中的图像提取及分割,[J].中文信息学报,2000,14(4):29-33.
[13]郁梅,郁伯康,郑义,基于视觉的车辆牌照检测,[J].计算机应用研究. 1999(5):65-67.
[14]张引,潘云鹤,彩色汽车图像牌照定位新方法,[J].中国图像图形学报. 2001,6(4):374-377.
[15]戚飞虎,叶芗芸,李卫东等,机动车辆自动收费系统,[J].交通与计算机,1997,15(1):38-41.
[16]戴青云,余英林,一种基于小波与形态学的车牌图像分割方法,[J].中国图形图像学报,2000,5A(5):411-415.
[17]万国红,王敏,黄心汉等,基于神经网络的汽车牌照自动识别技术研究,[J].计算机工程与应用,2002,(6):204-205.
[18]赵雪春,戚飞虎,基于彩色分割的车牌自动识别技术,[J].上海交通大学学报,1998,(10):3-9.
[19]廖金周,宣国荣,车辆牌照的自动分割,[J].微型电脑应用,1999(7):32-34.
[20]何斌,马天予,王运坚等,Visual C++数字图像处理,[M].北京:人民邮电出版社,2002.
[21] David J.Kruglinski, Scot Wingo, George Shepherd著,希望图书创作室译.ProgrammingVisual C++6.0技术内幕,[M].北京,北京希望电子出版社,2001.
[22]苏明金,刘宏,刘波,Matlab高级编程,电子工业出版社,2005:45-79.
[23]陈永超,李汉强,Matlab7.0与VC6.0的接口技术,电脑知识与技术,2006.1:129-130, SSN 1009-3044.
[24]朱志刚,数字图像处理基础,[M].北京,清华大学出版社,1996.
[25] Ostu N. A Threshold Selection Method from Gray-Level Histograms[J], IEEE Transactions, Systems, Man and Cybernetics, 1979, SMC-9(1): 62-66.
[26]王润生,图像理解,国防科技大学出版社,1995
[27]高朝晖,黄卫,基于彩色图像车牌分割研究,公路交通科技,2004,Vol.21, No.8:114-117.
[28]金玲玲,廖芹,汪刘一,汽车牌照的提取方法,华南理工大学学报,2002,Vo1.30 No.7:95-97.
[29]郭大波,陈礼民,卢朝阳等,基于车牌底色识别的车牌定位方法,计算机工程与设计,2004,Vol.24.No.124,No5:81-84.
[30]付仲良,黄书强,陈眠等.货车图像车牌区快速定位及字符切割算法.计算机工程与设计,2003, Vol.24 No.l:77-79.
[31] T Randen, J.H Husoy. Filtering for texture classification: a comparative study, IEEE Trans.Pattern Analysis and Machine Intelligence, 1999, Vo1.21 No.4.
[32]陈昕,基于纹理分析的车牌图像区域分割技术的研究,[硕士学位论文]合肥,合肥工业大学2003:17-32.
[33]汤德佑,黄元江,基于灰度跳变的车牌自动识别技术,株洲工学院学报,Mar. 2004 Vol. 18 No. 2: 1-2.
[34]熊军,高敦堂,沈庆宏等,基于字符纹理特征的快速定位算法,光电工程,2003, Vol. 30 No. 2: 60-63.
[35]林纯青,徐立亚,戚飞虎,汽车图象中字符目标的提取算法,上海交通大学学报,1998, Vol.33 No. 10:1-3.
[36]张西宁,郑南宁,汽车牌照自动识别中的目标与背景的快速分割,信息与控制,1998: 27-31.
[37]杨海廷,基于纹理特征的车牌识别系统的研究与实现,[硕士学位论文],成都,电子科技大学,2005:22-27.
[38] Barroso J, Rafael A. Number Plate Reading Using Computer Vision[J]. IEEE Transactionon Vehicular Technology. 2001.11.
[39] Rittscher C, Kato J. A Probabilistic Background Model for Tracking. Eccv 2000,Vol.2.
[40]宋建才,字符结构知识在车牌识别中的应用,电子技术应用,2004,Vol.30,No.5 :188-191.
[41]王建平,姜滔,基于小波分析的汽车牌照分割,合肥工业大学学报,2002 o1.25 No.6:1139-1142.
[42] Parist R. Car Plate Recognition by Neural Networks and Image Processing IEEE International Symposium on Circuits and Systems, USA, 2000, May 31-June 3.
[43]崔江,王友仁,车牌自动识别方法中的关键技术研究,[J].计算机测量与控制,2003,Vol.11 No.4:260-262.
[44] J.Bernsen,Dvnamic thresholdingof gay-level images,Proc.of 8th Int'l Conf. onPatt. Recongn,1986:1251-1255.
[45] Byeong Rae Lee, Kyungsoo Park, Hyunchul Kang ,et al. Adaptive Local Binarization Method for Recognition of Vehicle License Plates,IWCIA 2004, LNCS 3322, Springer-Verlag Berlin Heidelberg 2004:646–655.
[46]张洁玮,车牌识别系统中定位算法的研究,[硕士学位论文],南京,南京航空航天大学,2006.
[47]张铁弓,基于DSP的微光视频图像实时处理研究,[硕士学位论文],武汉,武汉理工大学,2002:33-36.
[48]胡小锋,赵辉,《Visual C++/MATLAB图像处理与识别实用案例精选》,[M].北京,人民邮电出版社,2004.
[49] T Randen, J.H Husoy. Filtering for texture classification: a comparative study, IEEE Trans.Pattern Analysis and Machine Intelligence, 1999, Vo1.21 No.4.
[50] Barroso J, Rafael A. Number Plate Reading Using Computer Vision[J]. IEEE Transaction on Vehicular Technology. 2001.11.
[51] Rittscher C, Kato J. A Probabilistic Background Model for Tracking. Eccv 2000,Vol.2.
[52]宋建才,字符结构知识在车牌识别中的应用,电子技术应用,2004,Vol.30,No.5 :188-191.
[53]王建平,姜滔,基于小波分析的汽车牌照分割,合肥工业大学学报,2002 o1.25 No.6:1139-1142.
[54]刘彦婷,徐海兰,杨磊,车牌识别软件的改进算法,北京广播学院学报(自然科学版), 2005年12月第12卷,第4期:39-43.
[55] David Llorens, Andres Marzal, et al. Car License Plates Extraction and Recognition Based on Connected Components Analysis and HMM Decoding[J]. IbPRIA 2005 , LNCS 3522,Springer-Verlag Berlin Heidelberg,2005:571–578.
[56]张云刚,张长水,利用Hough变换和先验知识的车牌字符分割算法,[J].计算机学报,2004.1:130-135.
[57]陈寅鹏,丁晓青,复杂车辆图像中的车牌定位与字符分割方法,[J].红外与激光工程,2004,33(1):29-33.
[58]卢雅琴,邬凌超。基于数学形态学的车牌定位方法。计算机工程,2005,2(31):224-225。
[2]蔡文沁,我国智能交通系统发展的战略构想.交通运输系统工程与信息,2003年2月,Vol 3, No.l:16-22.
[3] Javier Cano and Juan-Carlos P′erez-Cort′es F.J. Perales et al. (Eds.): Vehicle License Plate Segmentation in Natural Images, IbPRIA 2003, LNCS 2652, 2003._c Springer-Verlag Berlin Heidelberg 2003:142-149.
[4]尹令,车牌自动识别系统的研究及其在VC++中的实现,[硕士学位论文],湘潭,湘潭大学计算控制技术与控制理论应用,2003:1-3.
[5]陈旭梅,于雷,郭继孚等,美国智能交通系统ITS的近期发展综述,中外公路,2003 Vo1.23, No.2: 9-120.
[6] Barroso J,Bulas-Cruz J & Dagless E L,Real-Time Number Plate Reading[J]. 4th IFAC Workshop on Algorithms and Architectures for Real-Time Control,1997.
[7] R.Parisi. Car Plate Recognition by Neural Networks and Image Processing[J]. IEEE International Symposium on Industrial Electronics, 1998:605-610.
[8] C.H.Coetzee,et al. PC Based Number Plate Recognition Systems, In Proc.IEEE International Symposium on Industrial Electronics. 1999,1(1):605-610.
[9] C.Rittscher, J.Kato, S.Joga, and A.Blake., A Probabilistic Background Model for Tracking. 2000, 2(2): 63-67.
[10]刘效静,成瑜. ,汽车牌照自动识别技术研究,[J].南京航天航空大学学报, 1998,30(5):573-575.
[11]牛欣,沈兰荪,基于特征的车辆牌照定位算法,[J].交通与计算机,. 2001,1(81):31-33.
[12]刘智勇,刘迎建,车牌识别(LPR)中的图像提取及分割,[J].中文信息学报,2000,14(4):29-33.
[13]郁梅,郁伯康,郑义,基于视觉的车辆牌照检测,[J].计算机应用研究. 1999(5):65-67.
[14]张引,潘云鹤,彩色汽车图像牌照定位新方法,[J].中国图像图形学报. 2001,6(4):374-377.
[15]戚飞虎,叶芗芸,李卫东等,机动车辆自动收费系统,[J].交通与计算机,1997,15(1):38-41.
[16]戴青云,余英林,一种基于小波与形态学的车牌图像分割方法,[J].中国图形图像学报,2000,5A(5):411-415.
[17]万国红,王敏,黄心汉等,基于神经网络的汽车牌照自动识别技术研究,[J].计算机工程与应用,2002,(6):204-205.
[18]赵雪春,戚飞虎,基于彩色分割的车牌自动识别技术,[J].上海交通大学学报,1998,(10):3-9.
[19]廖金周,宣国荣,车辆牌照的自动分割,[J].微型电脑应用,1999(7):32-34.
[20]何斌,马天予,王运坚等,Visual C++数字图像处理,[M].北京:人民邮电出版社,2002.
[21] David J.Kruglinski, Scot Wingo, George Shepherd著,希望图书创作室译.ProgrammingVisual C++6.0技术内幕,[M].北京,北京希望电子出版社,2001.
[22]苏明金,刘宏,刘波,Matlab高级编程,电子工业出版社,2005:45-79.
[23]陈永超,李汉强,Matlab7.0与VC6.0的接口技术,电脑知识与技术,2006.1:129-130, SSN 1009-3044.
[24]朱志刚,数字图像处理基础,[M].北京,清华大学出版社,1996.
[25] Ostu N. A Threshold Selection Method from Gray-Level Histograms[J], IEEE Transactions, Systems, Man and Cybernetics, 1979, SMC-9(1): 62-66.
[26]王润生,图像理解,国防科技大学出版社,1995
[27]高朝晖,黄卫,基于彩色图像车牌分割研究,公路交通科技,2004,Vol.21, No.8:114-117.
[28]金玲玲,廖芹,汪刘一,汽车牌照的提取方法,华南理工大学学报,2002,Vo1.30 No.7:95-97.
[29]郭大波,陈礼民,卢朝阳等,基于车牌底色识别的车牌定位方法,计算机工程与设计,2004,Vol.24.No.124,No5:81-84.
[30]付仲良,黄书强,陈眠等.货车图像车牌区快速定位及字符切割算法.计算机工程与设计,2003, Vol.24 No.l:77-79.
[31] T Randen, J.H Husoy. Filtering for texture classification: a comparative study, IEEE Trans.Pattern Analysis and Machine Intelligence, 1999, Vo1.21 No.4.
[32]陈昕,基于纹理分析的车牌图像区域分割技术的研究,[硕士学位论文]合肥,合肥工业大学2003:17-32.
[33]汤德佑,黄元江,基于灰度跳变的车牌自动识别技术,株洲工学院学报,Mar. 2004 Vol. 18 No. 2: 1-2.
[34]熊军,高敦堂,沈庆宏等,基于字符纹理特征的快速定位算法,光电工程,2003, Vol. 30 No. 2: 60-63.
[35]林纯青,徐立亚,戚飞虎,汽车图象中字符目标的提取算法,上海交通大学学报,1998, Vol.33 No. 10:1-3.
[36]张西宁,郑南宁,汽车牌照自动识别中的目标与背景的快速分割,信息与控制,1998: 27-31.
[37]杨海廷,基于纹理特征的车牌识别系统的研究与实现,[硕士学位论文],成都,电子科技大学,2005:22-27.
[38] Barroso J, Rafael A. Number Plate Reading Using Computer Vision[J]. IEEE Transactionon Vehicular Technology. 2001.11.
[39] Rittscher C, Kato J. A Probabilistic Background Model for Tracking. Eccv 2000,Vol.2.
[40]宋建才,字符结构知识在车牌识别中的应用,电子技术应用,2004,Vol.30,No.5 :188-191.
[41]王建平,姜滔,基于小波分析的汽车牌照分割,合肥工业大学学报,2002 o1.25 No.6:1139-1142.
[42] Parist R. Car Plate Recognition by Neural Networks and Image Processing IEEE International Symposium on Circuits and Systems, USA, 2000, May 31-June 3.
[43]崔江,王友仁,车牌自动识别方法中的关键技术研究,[J].计算机测量与控制,2003,Vol.11 No.4:260-262.
[44] J.Bernsen,Dvnamic thresholdingof gay-level images,Proc.of 8th Int'l Conf. onPatt. Recongn,1986:1251-1255.
[45] Byeong Rae Lee, Kyungsoo Park, Hyunchul Kang ,et al. Adaptive Local Binarization Method for Recognition of Vehicle License Plates,IWCIA 2004, LNCS 3322, Springer-Verlag Berlin Heidelberg 2004:646–655.
[46]张洁玮,车牌识别系统中定位算法的研究,[硕士学位论文],南京,南京航空航天大学,2006.
[47]张铁弓,基于DSP的微光视频图像实时处理研究,[硕士学位论文],武汉,武汉理工大学,2002:33-36.
[48]胡小锋,赵辉,《Visual C++/MATLAB图像处理与识别实用案例精选》,[M].北京,人民邮电出版社,2004.
[49] T Randen, J.H Husoy. Filtering for texture classification: a comparative study, IEEE Trans.Pattern Analysis and Machine Intelligence, 1999, Vo1.21 No.4.
[50] Barroso J, Rafael A. Number Plate Reading Using Computer Vision[J]. IEEE Transaction on Vehicular Technology. 2001.11.
[51] Rittscher C, Kato J. A Probabilistic Background Model for Tracking. Eccv 2000,Vol.2.
[52]宋建才,字符结构知识在车牌识别中的应用,电子技术应用,2004,Vol.30,No.5 :188-191.
[53]王建平,姜滔,基于小波分析的汽车牌照分割,合肥工业大学学报,2002 o1.25 No.6:1139-1142.
[54]刘彦婷,徐海兰,杨磊,车牌识别软件的改进算法,北京广播学院学报(自然科学版), 2005年12月第12卷,第4期:39-43.
[55] David Llorens, Andres Marzal, et al. Car License Plates Extraction and Recognition Based on Connected Components Analysis and HMM Decoding[J]. IbPRIA 2005 , LNCS 3522,Springer-Verlag Berlin Heidelberg,2005:571–578.
[56]张云刚,张长水,利用Hough变换和先验知识的车牌字符分割算法,[J].计算机学报,2004.1:130-135.
[57]陈寅鹏,丁晓青,复杂车辆图像中的车牌定位与字符分割方法,[J].红外与激光工程,2004,33(1):29-33.
[58]卢雅琴,邬凌超。基于数学形态学的车牌定位方法。计算机工程,2005,2(31):224-225。