汽车牌照自动识别技术研究
|
摘要
车牌自动识别系统是以汽车牌照为特定目标的专用计算机视觉系统。一个典型车牌识别系统又分为车牌定位、车牌字符分割及字符识别三部分,它的研究主要涉及到了模式识别和人工智能、计算机视觉、数字图像处理、人工神经网络等众多的学科领域。
本课题的目的是对输入的汽车图像进行预处理,提取车牌区域,并对车牌字符进行分割,然后对字符逐个进行识别,最后按顺序以文本的形式表示出来,得到车牌号码的识别结果并将其保存在数据库中。在对车牌识别关键技术进行研究的基础上,完成了车牌识别系统软件的设计。
车牌定位是车牌识别系统处理的第一步,定位的准确与否直接关系着车牌识别的成败。首先对图像进行预处理,然后采用行扫描方法进行水平定位,再根据投影法确定车牌左右边界,在准确率上基本满足要求。
车牌字符的切分效果直接影响到字符识别。在字符切分之前,先对二值化处理后的车牌图像进行线扫描,确定字符纹理区域上、下边界线的水平倾斜角度,据此对车牌进行倾斜校正,并去除残余边框和铆钉。然后通过投影法确定字符左右边界,进行字符切分,并进行归一化处理。
车牌字符的识别是整个系统中最重要的一步,也是系统的设计目的所在。根据车牌字符的特点,采用BP神经网络进行分类识别。将归一化处理后的独立字符图像分别送入汉字识别网络、字母识别网络、字母数字识别网络和数字识别网络进行识别,最后把识别结果按原顺序组合,以文本形式输出,并存放到数据库中。
实验结果表明,本课题设计的车牌识别软件能比较准确地定位车牌区域、分割车牌字符并进行字符识别,且性能良好。
Automatic License Plate Recognition System is a specific computer vision system based on the specific goals of the vehicle license. A typical License Plate Recognition System is composed of License Plate Detection, Plate Character Segmentation and Character Recognition. The LPR system involves numerous discipline domains, such as Pattern Recognition and Artificial Intelligence, Computer Vision, Digital Image Processing, Artificial Neural Network etc.
The purpose of the paper is analyzed as follows. The image inputted to the computer is disposed and analyzed in order to recognize the characters on the license plate and complete the character segmentation and recognition. At last the vehicle license plate number is printed in text string according to the order of original images, and sent to the database. Based on the research of key technologies in LPR, a tentative LPR system is set up.
Locate the plate is the first step of the LPR system, whose accuracy directly affects the success of the character recognition result. Firstly, image pretreatment is processed. Secondly, row-scanning is adopted as the horizontal location method. Then, vertical projection is adopted to determine the left and right boundary of the plate. The result proved that, it can satisfy the demand in accuracy.
The result of character segmentation affects the recognition directly. Before character segmentation, line-scanning method is taken to determine the horizontal gradient of the top and bottom of character grain area. According to the angle, incline-revising is adopted, and at the same time remaining frame and rivets are also wiped off. Then, vertical projection is adopted to determine the left and right boundary of every character and segment the characters. After that, normalization is taken.
Character Recognition takes the most important part in the whole system, which is the purpose of the system design. According to the feature of characters on the plate, BP Artificial Neural Network is adopted as a classified recognition method. Single normalized character is delivered to Chinese character recognition sub-network, letter character recognition sub-network, letter and number recognition sub-network, and number recognition sub-network separately. Finally, the results are printed in text strings according to the order of original images, and sent to the database.
Experiment results show that, the LPR system discussed in the paper can accurately locate to license plate, segment and recognize characters. And the system's performance is good.
本课题的目的是对输入的汽车图像进行预处理,提取车牌区域,并对车牌字符进行分割,然后对字符逐个进行识别,最后按顺序以文本的形式表示出来,得到车牌号码的识别结果并将其保存在数据库中。在对车牌识别关键技术进行研究的基础上,完成了车牌识别系统软件的设计。
车牌定位是车牌识别系统处理的第一步,定位的准确与否直接关系着车牌识别的成败。首先对图像进行预处理,然后采用行扫描方法进行水平定位,再根据投影法确定车牌左右边界,在准确率上基本满足要求。
车牌字符的切分效果直接影响到字符识别。在字符切分之前,先对二值化处理后的车牌图像进行线扫描,确定字符纹理区域上、下边界线的水平倾斜角度,据此对车牌进行倾斜校正,并去除残余边框和铆钉。然后通过投影法确定字符左右边界,进行字符切分,并进行归一化处理。
车牌字符的识别是整个系统中最重要的一步,也是系统的设计目的所在。根据车牌字符的特点,采用BP神经网络进行分类识别。将归一化处理后的独立字符图像分别送入汉字识别网络、字母识别网络、字母数字识别网络和数字识别网络进行识别,最后把识别结果按原顺序组合,以文本形式输出,并存放到数据库中。
实验结果表明,本课题设计的车牌识别软件能比较准确地定位车牌区域、分割车牌字符并进行字符识别,且性能良好。
Automatic License Plate Recognition System is a specific computer vision system based on the specific goals of the vehicle license. A typical License Plate Recognition System is composed of License Plate Detection, Plate Character Segmentation and Character Recognition. The LPR system involves numerous discipline domains, such as Pattern Recognition and Artificial Intelligence, Computer Vision, Digital Image Processing, Artificial Neural Network etc.
The purpose of the paper is analyzed as follows. The image inputted to the computer is disposed and analyzed in order to recognize the characters on the license plate and complete the character segmentation and recognition. At last the vehicle license plate number is printed in text string according to the order of original images, and sent to the database. Based on the research of key technologies in LPR, a tentative LPR system is set up.
Locate the plate is the first step of the LPR system, whose accuracy directly affects the success of the character recognition result. Firstly, image pretreatment is processed. Secondly, row-scanning is adopted as the horizontal location method. Then, vertical projection is adopted to determine the left and right boundary of the plate. The result proved that, it can satisfy the demand in accuracy.
The result of character segmentation affects the recognition directly. Before character segmentation, line-scanning method is taken to determine the horizontal gradient of the top and bottom of character grain area. According to the angle, incline-revising is adopted, and at the same time remaining frame and rivets are also wiped off. Then, vertical projection is adopted to determine the left and right boundary of every character and segment the characters. After that, normalization is taken.
Character Recognition takes the most important part in the whole system, which is the purpose of the system design. According to the feature of characters on the plate, BP Artificial Neural Network is adopted as a classified recognition method. Single normalized character is delivered to Chinese character recognition sub-network, letter character recognition sub-network, letter and number recognition sub-network, and number recognition sub-network separately. Finally, the results are printed in text strings according to the order of original images, and sent to the database.
Experiment results show that, the LPR system discussed in the paper can accurately locate to license plate, segment and recognize characters. And the system's performance is good.
引文
[1]刘美生.我国机动车性能检测的现状与发展.中国测试技术,2005,31(6):1-5
[2]王淑玲,赵涛.车牌自动识别技术在汽车检测线上的应用.中国仪器仪表,2005,9:110-111
[3]林立,何为,韩力群.汽车牌照自动识别技术的现状与发展.北京轻工业学院学报,2001,19(1):36-40
[4]FaNny MMM.Computer Vision Application to Automatic Number-Plate Recognition.Proceedings of 26~(th)International Symposium on Automotive Technology and Automation,Aachen Germany.1993:625-633
[5]Jun Ohya,et al..Recognizing Character in Scene Images.IEEE fans,On Pattern Analysis and Machine Intelligence.1994,16(2):214-220
[6]Setchell J.Applications of Computer Vision to Road-Traffic Monitoring.PH.D Thesis University of Bristol,England.1997:66-81
[7]赵雪春,戚飞虎.基于彩色分割的车牌自动识别技术.上海交通大学学报,1998,32(10):4-9
[8]魏武,黄心汉,张起森,王敏,王明俊.基于模板匹配和神经网络的车牌字符识别方法.模式识别与人工智能,2001,14(1):123-127
[9]张美多,郭宝龙.车牌识别系统关键技术研究.计算机工程,2007,33(16):186-188
[10]李小平,林学訚,曲大成,任江兴.车辆牌照识别系统可靠性问题的研究.北京理工大学学报,2001,21(1):11-14
[11]胡硕.基于遗传算法的图像分割研究.长春:东北师范大学,2003
[12]刘耀辉.三种阈值计算方法在Matlab 6.5中的实现.湘南学院学报,2007,28(5):81-84
[13]陈丹,张峰,贺贵明.一种改进的文本图像二值化算法.计算机工程,2003,29(13):85-86
[14]蔡梅艳,吴庆宪,姜长生.改进Otsu法的目标图像分割.电光与控制,2007,14(6):118-119
[15]龚昌来.基于数据融合技术的图像均值滤波算法.微计算机信息,2007,23(6):313-314
[16]姜珊,王跃存.基于中值滤波和形态学的去噪算法.仪表仪器用户,2007,4(14):106-107
[17]董付国,原达,王金鹏.中值滤波快速算法的进一步思考.计算机工程与应用,2007, 43(26):48-49
[18]耿迅.VC图像处理——边缘检测.电脑编程技巧与维护,2006,2:63-68
[19]徐献灵,林奕水.图像边缘检测算法比较与分析.自动化与信息工程,2007(3):44-46
[20]刑军.基于Sobel算子数字图像的边缘检测.微机发展,2005,15(9):48-49
[21]刘明艳,赵景秀,孙宁.用Prewitt算子细化边缘.光电子技术,2006,26(4):261-263
[22]冯湘.图像分割的计算机实现.郑州铁路职业技术学院学报,2007,19(4):10-11
[23]万军,徐汀荣.基于Laplacian算子的图像边缘检测方法研究.现代电子技术,2004(21):92-93
[24]严国萍,何俊峰.高斯-拉普拉斯边缘检测算子的扩展研究.华中科技大学学报,2006,34(10):21-23
[25]蒋爱德,扈少华.基于Canny算子的边缘检测研究.郑州牧业工程高等专科学校学报,2007,27(2):38-40
[26]张震,马驷良,张忠波,刘辉,宫跃欣,孙秋成.一种基于Canny算子的图像边缘提取算法.吉林大学学报,2007,45(2):244-248
[27]刘彩.一种改进的Sobel图像边缘检测算法.贵州工业大学学报,2004,33(5):77-79
[28]刘效静,成瑜.汽车牌照自动识别技术研究.南京航空航天大学学报,1998,30(5):573-576
[29]袁志伟,潘晓露,陈艾,李一民.车辆牌照定位的算法研究.昆明理工大学学报,2001,26(2):56-59
[30]黎楷模.边缘检测算法在牌照定位中的应用分析.郑州工业高等专科学校学报,2004,20(3):1-2
[31]李刚,曾锐利,林凌,王蒙军.基于数学形态学的车牌定位方法.仪器仪表学报,2007,28(7):1323-1327
[32]黄豪杰,李榕,常鸿森,李南希.基于边缘颜色分布的车牌定位新方法.激光杂志,2007,28(3):57-59
[33]杨俊,戚飞虎.一种基于形状和纹理特征的车牌定位方法.计算机工程,2006,32(2):170-171
[34]陈建坤,范春年.一种基于神经网络的车牌定位方法.辽宁工程技术大学学报,2005,24(1):97-100
[35]刘丽新,刘京刚.行扫描进行车牌上下边界定位的研究.仪器仪表学报,2005,26(8):177-179
[36]应宏徽,姚明海,张永华.基于纹理分析和垂直投影的车牌定位算法.控制工程,2004,11(5):433-435
[37]芮挺,沈春林,张金林.车牌识别中倾斜牌照的快速矫正算法.计算机工程,2004,30(13):122-124
[38]路小波,包明,黄卫.基于投影的车牌倾斜检测方法.交通运输工程与信息学报,2004,2(4):10-15
[39]刘静,周静华,苏俊连,付佳.基于模板匹配的车牌字符识别算法实现.科技信息,2007(24):34-35
[40]Jianjun Ni,Xiaoping Ma,Lizhong Xu,Jianying Wang.An Image Recognition Method Based on Multiple BP Neural Networks Fusion.Proceedings of 2004 International Conference on Information Acquisition,2004:323-326
[41]R.Parisi,E.D.Di Claudio,G.Lucarelli,G..Orlandi.Car Plate Recognition by Neural Networks and Image Processing.In Proc.IEEE ISCAS.USA,1998:195-198
[42]单潮龙,马伟明,贲可荣,张磊.BP人工神经网络的应用及其实现技术.海军工程大学学报,2000(4):16-17
[43]Jinwei Wen,Jiali Zhao,Siwei Luo,Zhen Han.The Improvements of BP Neural Network Learning Algorithm.Proceedings of ICSP.2000:1647-1649
[44]叶良,蒋凯男,高建民,刘仲良.VC中的ADO开发简介.计算机时代,2002(8):20-21
[45]徐鑫,康波,吕炳朝.基于ADO的数据库编程技术在VC++中的应用.微机发展,2004,12(14):92-94
[2]王淑玲,赵涛.车牌自动识别技术在汽车检测线上的应用.中国仪器仪表,2005,9:110-111
[3]林立,何为,韩力群.汽车牌照自动识别技术的现状与发展.北京轻工业学院学报,2001,19(1):36-40
[4]FaNny MMM.Computer Vision Application to Automatic Number-Plate Recognition.Proceedings of 26~(th)International Symposium on Automotive Technology and Automation,Aachen Germany.1993:625-633
[5]Jun Ohya,et al..Recognizing Character in Scene Images.IEEE fans,On Pattern Analysis and Machine Intelligence.1994,16(2):214-220
[6]Setchell J.Applications of Computer Vision to Road-Traffic Monitoring.PH.D Thesis University of Bristol,England.1997:66-81
[7]赵雪春,戚飞虎.基于彩色分割的车牌自动识别技术.上海交通大学学报,1998,32(10):4-9
[8]魏武,黄心汉,张起森,王敏,王明俊.基于模板匹配和神经网络的车牌字符识别方法.模式识别与人工智能,2001,14(1):123-127
[9]张美多,郭宝龙.车牌识别系统关键技术研究.计算机工程,2007,33(16):186-188
[10]李小平,林学訚,曲大成,任江兴.车辆牌照识别系统可靠性问题的研究.北京理工大学学报,2001,21(1):11-14
[11]胡硕.基于遗传算法的图像分割研究.长春:东北师范大学,2003
[12]刘耀辉.三种阈值计算方法在Matlab 6.5中的实现.湘南学院学报,2007,28(5):81-84
[13]陈丹,张峰,贺贵明.一种改进的文本图像二值化算法.计算机工程,2003,29(13):85-86
[14]蔡梅艳,吴庆宪,姜长生.改进Otsu法的目标图像分割.电光与控制,2007,14(6):118-119
[15]龚昌来.基于数据融合技术的图像均值滤波算法.微计算机信息,2007,23(6):313-314
[16]姜珊,王跃存.基于中值滤波和形态学的去噪算法.仪表仪器用户,2007,4(14):106-107
[17]董付国,原达,王金鹏.中值滤波快速算法的进一步思考.计算机工程与应用,2007, 43(26):48-49
[18]耿迅.VC图像处理——边缘检测.电脑编程技巧与维护,2006,2:63-68
[19]徐献灵,林奕水.图像边缘检测算法比较与分析.自动化与信息工程,2007(3):44-46
[20]刑军.基于Sobel算子数字图像的边缘检测.微机发展,2005,15(9):48-49
[21]刘明艳,赵景秀,孙宁.用Prewitt算子细化边缘.光电子技术,2006,26(4):261-263
[22]冯湘.图像分割的计算机实现.郑州铁路职业技术学院学报,2007,19(4):10-11
[23]万军,徐汀荣.基于Laplacian算子的图像边缘检测方法研究.现代电子技术,2004(21):92-93
[24]严国萍,何俊峰.高斯-拉普拉斯边缘检测算子的扩展研究.华中科技大学学报,2006,34(10):21-23
[25]蒋爱德,扈少华.基于Canny算子的边缘检测研究.郑州牧业工程高等专科学校学报,2007,27(2):38-40
[26]张震,马驷良,张忠波,刘辉,宫跃欣,孙秋成.一种基于Canny算子的图像边缘提取算法.吉林大学学报,2007,45(2):244-248
[27]刘彩.一种改进的Sobel图像边缘检测算法.贵州工业大学学报,2004,33(5):77-79
[28]刘效静,成瑜.汽车牌照自动识别技术研究.南京航空航天大学学报,1998,30(5):573-576
[29]袁志伟,潘晓露,陈艾,李一民.车辆牌照定位的算法研究.昆明理工大学学报,2001,26(2):56-59
[30]黎楷模.边缘检测算法在牌照定位中的应用分析.郑州工业高等专科学校学报,2004,20(3):1-2
[31]李刚,曾锐利,林凌,王蒙军.基于数学形态学的车牌定位方法.仪器仪表学报,2007,28(7):1323-1327
[32]黄豪杰,李榕,常鸿森,李南希.基于边缘颜色分布的车牌定位新方法.激光杂志,2007,28(3):57-59
[33]杨俊,戚飞虎.一种基于形状和纹理特征的车牌定位方法.计算机工程,2006,32(2):170-171
[34]陈建坤,范春年.一种基于神经网络的车牌定位方法.辽宁工程技术大学学报,2005,24(1):97-100
[35]刘丽新,刘京刚.行扫描进行车牌上下边界定位的研究.仪器仪表学报,2005,26(8):177-179
[36]应宏徽,姚明海,张永华.基于纹理分析和垂直投影的车牌定位算法.控制工程,2004,11(5):433-435
[37]芮挺,沈春林,张金林.车牌识别中倾斜牌照的快速矫正算法.计算机工程,2004,30(13):122-124
[38]路小波,包明,黄卫.基于投影的车牌倾斜检测方法.交通运输工程与信息学报,2004,2(4):10-15
[39]刘静,周静华,苏俊连,付佳.基于模板匹配的车牌字符识别算法实现.科技信息,2007(24):34-35
[40]Jianjun Ni,Xiaoping Ma,Lizhong Xu,Jianying Wang.An Image Recognition Method Based on Multiple BP Neural Networks Fusion.Proceedings of 2004 International Conference on Information Acquisition,2004:323-326
[41]R.Parisi,E.D.Di Claudio,G.Lucarelli,G..Orlandi.Car Plate Recognition by Neural Networks and Image Processing.In Proc.IEEE ISCAS.USA,1998:195-198
[42]单潮龙,马伟明,贲可荣,张磊.BP人工神经网络的应用及其实现技术.海军工程大学学报,2000(4):16-17
[43]Jinwei Wen,Jiali Zhao,Siwei Luo,Zhen Han.The Improvements of BP Neural Network Learning Algorithm.Proceedings of ICSP.2000:1647-1649
[44]叶良,蒋凯男,高建民,刘仲良.VC中的ADO开发简介.计算机时代,2002(8):20-21
[45]徐鑫,康波,吕炳朝.基于ADO的数据库编程技术在VC++中的应用.微机发展,2004,12(14):92-94
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |