复杂背景下的车牌识别技术研究
|
摘要
车辆牌照识别是智能交通系统(ITS)的一个重要组成部分,尤其是复杂背景下的车牌识别,有着广泛的应用领域和美好的应用前景,其发展必将大大加速ITS进程。本文针对目前车牌识别算法的局限性,主要研究了复杂背景下的车牌识别技术。
本文主要从以下几个方面进行研究:1)复杂背景下的车体检测和定位。本文采用瞬时块差分法来检测车体,通过对视频序列中连续两帧图像做以点为中心的3×3框架的块差分,并取差分的平方以加强运动信息,然后利用滤波、膨胀等技术准确定位车体。实验表明该算法具有很好的检测效果。2)车牌定位。先对车辆图像进行二值化并进行一系列数学形态学操作,再利用车牌先验信息进行过滤,产生车牌候选区域。然后对候选车牌区域进行彩色边缘检测,利用车牌区域固定的颜色对信息和结构信息准确定位车牌。算法在满足实时性的同时拥有极高的定位准确率。3)车牌字符分割。由于车牌区域除了汉字之外,其它的字符都是连通的,本文采用基于纹理信息和连通区域相结合的方法来提取单个字符。算法实现简单,而且可以在一定程度上消除预处理效果较差带来的影响,同时又可从根本上解决了倾斜车牌分割的问题。4)车牌字符识别。提出了基于Null Space PCA和改进的BP神经网络相结合的字符识别方法。首先对预处理过的字符图像进行Null Space PCA变换以达到降维的目的,然后把不同位的车牌字符分别放入汉字、字母、字母数字和数字四个不同的神经网络分类器,用改进的BP算法进行训练识别。实验结果表明该算法具有较高的识别率和识别速度,能够满足实时识别的要求。
本文结合理论和实践,对复杂背景下的车牌识别系统的各关键技术进行了深入研究,并提出了新的思想和方法,对车牌字符识别技术的研究具有很大的意义和应用价值。
License Plate Recognition is one important constituent part of the Intelligence Transportation System (ITS). Especially it is the license plate recognition in complicated contexts. It covers comprehensive application areas and enjoys a great application perspective. Its development is sure to accelerate the progress of ITS. In light of the weak points in contemporary license plate recognition algorithm, the present study focuses on the techniques of license plate recognition within complicated contexts.
This thesis is divided into the following several aspects for the research: 1) the detection and location of the vehicle in complicated contexts. The present study employs the instantaneous difference method based on block, that is, to detect the vehicle by doing the block difference of the frame 3×3 with dots in the center for two successive images in the video array and then squaring the difference to strengthen the information of movement. The vehicle is accurately located with the help of techniques such as filter and dilation. The experiment has showed that this kind of algorithm holds sound testing effects. 2) license plate recognition. The first step is to make binary images of vehicles and to do a series of mathematic morphology operation. And then is to filter according to the existing information and to come up with the license plate areas for candidacy. After that, the color edge detection is made for the chosen license plate areas and license plates are accurately located according to fixed color-pair information and structural information in license plate areas. This kind of algorithm can reach the accurate location rate with a rather high degree as well as satisfy the real-time performance. 3) character segmentation of license plates. Because characters in the license plate areas are connected except for Chinese characters, the present study abstracts single characters by the way of combining information based on veins and connected domains. It is easy to do the algorithm, and to some extent, this can help eliminate the bad effect brought about by pretreatment as well as fundamentally solve the problem of slant license plates' segmentation. 4) character recognition of license plates. This part puts forward a new way of character recognition which is based on Null Space PCA and the improved BP neural network. At first, for the aim of dimension, the pretreated character images should be made to undergo the transfer of Null Space PCA. Then different characters should be put respectively into four different classifiers of neural network (they are Chinese characters, letters, letters and numbers, and numbers). The result of the experiment shows that this kind of algorithm can reach the high recognition rate and recognition speed, which can satisfy the demand of recognition in the real-time performance.
The present study combines both theories and practice, and delves into the various key techniques in the license plate recognition system under complicated contexts. New ideas and methods are given, which has great research significance and adds to the value of application for the character recognition technique of license plates.
本文主要从以下几个方面进行研究:1)复杂背景下的车体检测和定位。本文采用瞬时块差分法来检测车体,通过对视频序列中连续两帧图像做以点为中心的3×3框架的块差分,并取差分的平方以加强运动信息,然后利用滤波、膨胀等技术准确定位车体。实验表明该算法具有很好的检测效果。2)车牌定位。先对车辆图像进行二值化并进行一系列数学形态学操作,再利用车牌先验信息进行过滤,产生车牌候选区域。然后对候选车牌区域进行彩色边缘检测,利用车牌区域固定的颜色对信息和结构信息准确定位车牌。算法在满足实时性的同时拥有极高的定位准确率。3)车牌字符分割。由于车牌区域除了汉字之外,其它的字符都是连通的,本文采用基于纹理信息和连通区域相结合的方法来提取单个字符。算法实现简单,而且可以在一定程度上消除预处理效果较差带来的影响,同时又可从根本上解决了倾斜车牌分割的问题。4)车牌字符识别。提出了基于Null Space PCA和改进的BP神经网络相结合的字符识别方法。首先对预处理过的字符图像进行Null Space PCA变换以达到降维的目的,然后把不同位的车牌字符分别放入汉字、字母、字母数字和数字四个不同的神经网络分类器,用改进的BP算法进行训练识别。实验结果表明该算法具有较高的识别率和识别速度,能够满足实时识别的要求。
本文结合理论和实践,对复杂背景下的车牌识别系统的各关键技术进行了深入研究,并提出了新的思想和方法,对车牌字符识别技术的研究具有很大的意义和应用价值。
License Plate Recognition is one important constituent part of the Intelligence Transportation System (ITS). Especially it is the license plate recognition in complicated contexts. It covers comprehensive application areas and enjoys a great application perspective. Its development is sure to accelerate the progress of ITS. In light of the weak points in contemporary license plate recognition algorithm, the present study focuses on the techniques of license plate recognition within complicated contexts.
This thesis is divided into the following several aspects for the research: 1) the detection and location of the vehicle in complicated contexts. The present study employs the instantaneous difference method based on block, that is, to detect the vehicle by doing the block difference of the frame 3×3 with dots in the center for two successive images in the video array and then squaring the difference to strengthen the information of movement. The vehicle is accurately located with the help of techniques such as filter and dilation. The experiment has showed that this kind of algorithm holds sound testing effects. 2) license plate recognition. The first step is to make binary images of vehicles and to do a series of mathematic morphology operation. And then is to filter according to the existing information and to come up with the license plate areas for candidacy. After that, the color edge detection is made for the chosen license plate areas and license plates are accurately located according to fixed color-pair information and structural information in license plate areas. This kind of algorithm can reach the accurate location rate with a rather high degree as well as satisfy the real-time performance. 3) character segmentation of license plates. Because characters in the license plate areas are connected except for Chinese characters, the present study abstracts single characters by the way of combining information based on veins and connected domains. It is easy to do the algorithm, and to some extent, this can help eliminate the bad effect brought about by pretreatment as well as fundamentally solve the problem of slant license plates' segmentation. 4) character recognition of license plates. This part puts forward a new way of character recognition which is based on Null Space PCA and the improved BP neural network. At first, for the aim of dimension, the pretreated character images should be made to undergo the transfer of Null Space PCA. Then different characters should be put respectively into four different classifiers of neural network (they are Chinese characters, letters, letters and numbers, and numbers). The result of the experiment shows that this kind of algorithm can reach the high recognition rate and recognition speed, which can satisfy the demand of recognition in the real-time performance.
The present study combines both theories and practice, and delves into the various key techniques in the license plate recognition system under complicated contexts. New ideas and methods are given, which has great research significance and adds to the value of application for the character recognition technique of license plates.
引文
[1]Kameda Yoshinari,Minoh Michihiko.A Human Motion Estimation Method Using 3-successive Video Frames[C].Proceedings of International Conference on Virtual Systems and Multimedia.1996:135-140.
[2]Paragios N.,Deriche R..Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects[J].In:Proc.IEEE Transactions on Pattern Analysis and Machine Intelligence,2000.22(3):266-280.
[3]Takaba S..A Traffic Flow Measuring System Using a Solid State Sensor[C].In:Proc.IEEE Conf on Road Traffic Data Collection,London,UK,1984.45-56.
[4]Hoose N..An image Analysis Tool for Motor Way analysis and Surveillance[J].Traffic Engineering Control,1992,23(4):140-147.
[5]Kottle D.,Sun Y..Motion Estimation via Cluster Matching[J].In:Proc.IEEE Transactions on Pattern Analysis and Machine Intelligence,1994.16(2):1128-1132.
[6]Badenas J.,Bober M.,Pia E.Motion and Intensity-based Segmentation and Its Application to Traffic Monitoring[C].Proceedings of the International Conference on Image Analysis and Proceedings ICIAP'97,Florence,Italy,IAPR,1997:502-509.
[7]Barroso J.,Rafael A..Number Plate Reading Using Computer Vision.In:Proc.IEEE International Symposium on Industrial Electronics(ISLE).Portugal,http://www.utad.pt/,1997.
[8]Parisi R.Car Plate Recognition by Neural Networks and Image Processing.In:Proc.IEEE International Symposium on Circuits and Systems,USA,1998,May 31-June 3.
[9]Barroso J.,Bulas Cruz J,Dagless E L.Real-Time Number Plate Reading.IFAC Workshop on Algorithms for Real-Time Control,http://www.utad.pt/,1997.
[10]Charl Coetzee.PC Based Number Plate Recognition Systems.In:Proc.IEEE International Symposium on Industrial Electronics.1998,605-610.
[11]刘效静,成瑜.汽车牌照自动识别技术研究[J].南京航空航天大学学报,1998.30(5):573-576.
[12]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,2000.1(18):31-33.
[13]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.
[14]郁梅,郁伯康,郑义.基于视觉的乍辆牌照检测[J].计算机应用研究,1998.16(5):65-67.
[15]戚飞虎,叶萝芸.机动车辆自动识别收费系统工程[J].交通与计算机,1997.16(5):38-41.
[16]王昱,赵正校,杨硕.一种基于样本线搜索和霍夫变换的区域定位方法[J].红外线与激光工程,1999.28(4):16-21.
[17]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,1998.32(10):4-9.
[18]施鹏飞,郭捷.基于颜色和纹理分析的车牌定位方法[J].中国图像图形学报,2002.
[19]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,2001.10.
[20]叶晨洲,廖金周,梅帆.车辆牌照字符识别系统[J].计算机应用与研究,1999(7):85-87.
[21]曹迪铭,宣国荣.牌照字符分割中的区域分裂与合并[J].计算机工程,2000.26(10):18-20.
[22]任明亮,范勇,蒋欣荣等.一种伙速精确的汽车牌照字符切分算法[J].四川大学学报 (自然科学版),2002.39(3):450-454.
[23]Ohya J.,Shio A.,Akamatsu S.Recognizing Characters in Scene Images[J].Pattern Analysis and Machine Intelligence,IEEE Transactions on,1994.16:214-220.
[24]黄可为,赵雁南,孙发杰等.一种高效目标搜索方法及其在图象分析中的应用[J].中国图象图形学报,1999.12(4):1038-1043.
[25]魏武,张起森,王明俊等.一种基于模板匹配的车牌识别方法[J].中国公路学报,2001.14(1):104-106.
[26]陈斌,游志胜.车牌号码松弛模板匹配方法[J].计算机应用,2001.21(8):154-157.
[27]马俊利,莫玉龙,王明祥.一种基于改进模板匹配的车牌字符识别方法[J].小型微型计算机系统,2003.24(9):1670-1672.
[28]Tindall T W.Application of Neural Network Technique to Automatic License Plate Recognition[A].In:Proceedings of European Convention on Security and Detection[C],Brighton,England,1995:81-85.
[29]沈会良,李志能.基于矩和小波变换的数字、字母字符识别研究[J].中国图像图形学报(A版),2000.5(3):249-52.
[30]冯国进,顾国华.车牌自动定位与模糊识别算法[J].光电子·激光,2003.14(7):749-752.
[31]黄志斌,陈锻生.支持向量机在车牌字符识别中的应用[J].计算机工程,2003.29(5):192-194.
[32]梁玮,贾云得.一种复杂背景下的多车牌分割与识别方法[J].北京理工大学学报,2003.23(1):91-94
[33]李文举.智能交通中图像处理技术应用的研究[D].大连海事大学.2004.
[34]LeBourgeo E Robust Multi-font OCR System from Gray Level Images[Z].ICDAR Ulm,1997.
[35]胡新荣.一种改进的车牌字符分割算法[J].计算机工程,3006.32(3):191-193
[36]刘关松,陆宗骐,徐建国等.几种彩色模型在不同光照条件下的稳定性分析.小型微型计算机系统,2002.23(7):882-885
[37]Jenn-Kwei Tyan,Claus Neubaner.Character Segmentation Method for Vehicle License Plate Recognition|P].United State.US6473417B1.2002.10.
[38]戴青云,余英林.数学形态学在图像处理中的应用进展,控制理论与应用,2001.18(4):478-482
[39]Stromberg J O.A modified Franklin System and Higher Order Spline Systems on R" as Unconditional Bases for Hardy Spaces[J].In:Beckner W.Conf in honor of A Zygmund.New York:Academic Press,1986,475-493
[40]左奇,史忠科.一种基于数学形态学的实时车牌图像分割方法.中国图象图形学报,2003.8(3):281-285.
[41]蔡元明.神经网络识别手写体数字预处理后样本空间凸集性研究[D].中国科学院半导体研究所.
[42]刘滨.基于神经网络的车牌字符识别研究[D].武汉人学.2004
[43]江炜亮.车牌图像自动定位与识别算法的研究[D].国防科学技术人学.2003
[44]Namrata Vaswani,Rama Chellappa.Principal Component Null Space Analysis for Image/Video Classification[J].In:Proc.IEEE Transactions on Image Processing.http://citeseer.ist.psu.edu/678098.html.2004.
[2]Paragios N.,Deriche R..Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects[J].In:Proc.IEEE Transactions on Pattern Analysis and Machine Intelligence,2000.22(3):266-280.
[3]Takaba S..A Traffic Flow Measuring System Using a Solid State Sensor[C].In:Proc.IEEE Conf on Road Traffic Data Collection,London,UK,1984.45-56.
[4]Hoose N..An image Analysis Tool for Motor Way analysis and Surveillance[J].Traffic Engineering Control,1992,23(4):140-147.
[5]Kottle D.,Sun Y..Motion Estimation via Cluster Matching[J].In:Proc.IEEE Transactions on Pattern Analysis and Machine Intelligence,1994.16(2):1128-1132.
[6]Badenas J.,Bober M.,Pia E.Motion and Intensity-based Segmentation and Its Application to Traffic Monitoring[C].Proceedings of the International Conference on Image Analysis and Proceedings ICIAP'97,Florence,Italy,IAPR,1997:502-509.
[7]Barroso J.,Rafael A..Number Plate Reading Using Computer Vision.In:Proc.IEEE International Symposium on Industrial Electronics(ISLE).Portugal,http://www.utad.pt/,1997.
[8]Parisi R.Car Plate Recognition by Neural Networks and Image Processing.In:Proc.IEEE International Symposium on Circuits and Systems,USA,1998,May 31-June 3.
[9]Barroso J.,Bulas Cruz J,Dagless E L.Real-Time Number Plate Reading.IFAC Workshop on Algorithms for Real-Time Control,http://www.utad.pt/,1997.
[10]Charl Coetzee.PC Based Number Plate Recognition Systems.In:Proc.IEEE International Symposium on Industrial Electronics.1998,605-610.
[11]刘效静,成瑜.汽车牌照自动识别技术研究[J].南京航空航天大学学报,1998.30(5):573-576.
[12]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,2000.1(18):31-33.
[13]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.
[14]郁梅,郁伯康,郑义.基于视觉的乍辆牌照检测[J].计算机应用研究,1998.16(5):65-67.
[15]戚飞虎,叶萝芸.机动车辆自动识别收费系统工程[J].交通与计算机,1997.16(5):38-41.
[16]王昱,赵正校,杨硕.一种基于样本线搜索和霍夫变换的区域定位方法[J].红外线与激光工程,1999.28(4):16-21.
[17]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,1998.32(10):4-9.
[18]施鹏飞,郭捷.基于颜色和纹理分析的车牌定位方法[J].中国图像图形学报,2002.
[19]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,2001.10.
[20]叶晨洲,廖金周,梅帆.车辆牌照字符识别系统[J].计算机应用与研究,1999(7):85-87.
[21]曹迪铭,宣国荣.牌照字符分割中的区域分裂与合并[J].计算机工程,2000.26(10):18-20.
[22]任明亮,范勇,蒋欣荣等.一种伙速精确的汽车牌照字符切分算法[J].四川大学学报 (自然科学版),2002.39(3):450-454.
[23]Ohya J.,Shio A.,Akamatsu S.Recognizing Characters in Scene Images[J].Pattern Analysis and Machine Intelligence,IEEE Transactions on,1994.16:214-220.
[24]黄可为,赵雁南,孙发杰等.一种高效目标搜索方法及其在图象分析中的应用[J].中国图象图形学报,1999.12(4):1038-1043.
[25]魏武,张起森,王明俊等.一种基于模板匹配的车牌识别方法[J].中国公路学报,2001.14(1):104-106.
[26]陈斌,游志胜.车牌号码松弛模板匹配方法[J].计算机应用,2001.21(8):154-157.
[27]马俊利,莫玉龙,王明祥.一种基于改进模板匹配的车牌字符识别方法[J].小型微型计算机系统,2003.24(9):1670-1672.
[28]Tindall T W.Application of Neural Network Technique to Automatic License Plate Recognition[A].In:Proceedings of European Convention on Security and Detection[C],Brighton,England,1995:81-85.
[29]沈会良,李志能.基于矩和小波变换的数字、字母字符识别研究[J].中国图像图形学报(A版),2000.5(3):249-52.
[30]冯国进,顾国华.车牌自动定位与模糊识别算法[J].光电子·激光,2003.14(7):749-752.
[31]黄志斌,陈锻生.支持向量机在车牌字符识别中的应用[J].计算机工程,2003.29(5):192-194.
[32]梁玮,贾云得.一种复杂背景下的多车牌分割与识别方法[J].北京理工大学学报,2003.23(1):91-94
[33]李文举.智能交通中图像处理技术应用的研究[D].大连海事大学.2004.
[34]LeBourgeo E Robust Multi-font OCR System from Gray Level Images[Z].ICDAR Ulm,1997.
[35]胡新荣.一种改进的车牌字符分割算法[J].计算机工程,3006.32(3):191-193
[36]刘关松,陆宗骐,徐建国等.几种彩色模型在不同光照条件下的稳定性分析.小型微型计算机系统,2002.23(7):882-885
[37]Jenn-Kwei Tyan,Claus Neubaner.Character Segmentation Method for Vehicle License Plate Recognition|P].United State.US6473417B1.2002.10.
[38]戴青云,余英林.数学形态学在图像处理中的应用进展,控制理论与应用,2001.18(4):478-482
[39]Stromberg J O.A modified Franklin System and Higher Order Spline Systems on R" as Unconditional Bases for Hardy Spaces[J].In:Beckner W.Conf in honor of A Zygmund.New York:Academic Press,1986,475-493
[40]左奇,史忠科.一种基于数学形态学的实时车牌图像分割方法.中国图象图形学报,2003.8(3):281-285.
[41]蔡元明.神经网络识别手写体数字预处理后样本空间凸集性研究[D].中国科学院半导体研究所.
[42]刘滨.基于神经网络的车牌字符识别研究[D].武汉人学.2004
[43]江炜亮.车牌图像自动定位与识别算法的研究[D].国防科学技术人学.2003
[44]Namrata Vaswani,Rama Chellappa.Principal Component Null Space Analysis for Image/Video Classification[J].In:Proc.IEEE Transactions on Image Processing.http://citeseer.ist.psu.edu/678098.html.2004.