复杂场景下的车牌检测算法研究及其工程实现
|
摘要
车牌识别系统(LPR)是智能交通的重要组成部分,在现代交通管理、社会治安等方面发挥着重要作用,有着广泛的应用前景。车牌检测是车牌识别系统的关键技术之一。随着车牌识别系统的广泛使用,过去在受控环境和简单场景下的车牌检测算法已经不能满足当前的需要。因此,在复杂场景和成像条件情况下的车牌检测逐渐成为当前研究的热点。
本文对复杂场景和成像条件下的车牌检测算法进行了研究,主要完成了以下工作。
首先,本文实现并改进了一种启发式的车牌检测算法。该算法根据车牌区域垂直边缘丰富的特点,提取图像中的垂直边缘,并通过连接边缘,获得车牌候选区域。最后根据车牌的几何、边缘分布以及颜色等特征对候选区域进行筛选得到车牌区域。
其次,本文实现了一种基于Adaboost机器学习算法的车牌检测算法。在该算法中,定义了一种表征局部灰度与全局灰度比例的矩形特征,使用CS-Adaboost算法训练了一个由85个特征组成的级联分类器,并由此实现了一种车牌检测算法。以上两种算法都在PC平台上编程实现,并且在一个583张图像组成的测试集上进行了实验,给出了结果和性能分析。
此外,图像中的倾斜车牌不利于后续的字符分割等操作。针对该问题,本文实现了一种基于“跨栏模型”和“窄孔透射模型”的倾斜车牌矫正算法,给出了实验结果。
最后,本文将启发式车牌检测算法移植到了DSP平台上,构建了一个车牌检测系统,可以在CIF格式的图像上达到10f/s的检测速度。文中介绍了算法在DSP平台上的移植和优化过程,并给出了实验结果。
License Plate Recognition (LPR) system is an important part of the Intelligent Transportation System (ITS), and plays an important role in modern traffic management and social security. License Plate Detection is one of the key technologies in LPR systems. With the extensive application of LPR systems, the algorithms used to work under controlled conditions and simple scenes is unable to fulfill the requirement. Therefore, license plate detection under complex scenes and imaging conditions becoming a research focus.
In this thesis, license plate detection under complex scenes and imaging conditions is investigated, contributions are listed below.
First, an improved heuristic license plate detection algorithm has been realized. In this algorithm, edge extraction is performed first according to the characteristics of regions contain license plate that they have a high density of edge information. By connecting the edges, several candidates are generated. The license plate was obtained by filtering these candidates with geometry features, edge distribution and color.
Second, an algorithm based on Adaboost has been implemented. A kind of rectangle feature is selected, which indicates the percentage of the sum of a local area in the whole region. A cascade classifier with 85 features was trained by CS-Adaboost algorithm, and then a license plate detection algorithm was implemented. An experiment on a test set contains 583 images is carried out, the result and performance analysis was presented.
Since the slant of the license plate would affect the character segmentation and other subsequent operation, a slant correction algorithm was implemented based on the hurdle model and tubiform interspace projection model. The result of experiment was presented.
Finally, a license plate detection system was implemented by transplant the heuristic algorithm to DSP platform. This system can perform license plate detection on CIF images at the speed of 10 f/s. The result of experiment was reported.
本文对复杂场景和成像条件下的车牌检测算法进行了研究,主要完成了以下工作。
首先,本文实现并改进了一种启发式的车牌检测算法。该算法根据车牌区域垂直边缘丰富的特点,提取图像中的垂直边缘,并通过连接边缘,获得车牌候选区域。最后根据车牌的几何、边缘分布以及颜色等特征对候选区域进行筛选得到车牌区域。
其次,本文实现了一种基于Adaboost机器学习算法的车牌检测算法。在该算法中,定义了一种表征局部灰度与全局灰度比例的矩形特征,使用CS-Adaboost算法训练了一个由85个特征组成的级联分类器,并由此实现了一种车牌检测算法。以上两种算法都在PC平台上编程实现,并且在一个583张图像组成的测试集上进行了实验,给出了结果和性能分析。
此外,图像中的倾斜车牌不利于后续的字符分割等操作。针对该问题,本文实现了一种基于“跨栏模型”和“窄孔透射模型”的倾斜车牌矫正算法,给出了实验结果。
最后,本文将启发式车牌检测算法移植到了DSP平台上,构建了一个车牌检测系统,可以在CIF格式的图像上达到10f/s的检测速度。文中介绍了算法在DSP平台上的移植和优化过程,并给出了实验结果。
License Plate Recognition (LPR) system is an important part of the Intelligent Transportation System (ITS), and plays an important role in modern traffic management and social security. License Plate Detection is one of the key technologies in LPR systems. With the extensive application of LPR systems, the algorithms used to work under controlled conditions and simple scenes is unable to fulfill the requirement. Therefore, license plate detection under complex scenes and imaging conditions becoming a research focus.
In this thesis, license plate detection under complex scenes and imaging conditions is investigated, contributions are listed below.
First, an improved heuristic license plate detection algorithm has been realized. In this algorithm, edge extraction is performed first according to the characteristics of regions contain license plate that they have a high density of edge information. By connecting the edges, several candidates are generated. The license plate was obtained by filtering these candidates with geometry features, edge distribution and color.
Second, an algorithm based on Adaboost has been implemented. A kind of rectangle feature is selected, which indicates the percentage of the sum of a local area in the whole region. A cascade classifier with 85 features was trained by CS-Adaboost algorithm, and then a license plate detection algorithm was implemented. An experiment on a test set contains 583 images is carried out, the result and performance analysis was presented.
Since the slant of the license plate would affect the character segmentation and other subsequent operation, a slant correction algorithm was implemented based on the hurdle model and tubiform interspace projection model. The result of experiment was presented.
Finally, a license plate detection system was implemented by transplant the heuristic algorithm to DSP platform. This system can perform license plate detection on CIF images at the speed of 10 f/s. The result of experiment was reported.
引文
1 刘丽英,潘景山,史永,等.智能交通系统的发展状况研究.信息技术与信息化.2005.6:17-18
2 Barroso J,Dagless E L,Rafael A,et al.Number plate reading using computer vision.IEEE International Symposium on Industrial Electronics.1997,3:761-766
3 Parisi R,Di Claudio E D,Lucarelli G,et al.Car Plate Recognition by Neural Networks and Image Processing.IEEE Intemational Symposium on Circuits and Systems.1998,3:195-198
4 Coetzee C,Botha C,Weber D.PC Based Number Plate Recognition System.In Proc.IEEE International Symposium on Industrial Eleetronies.1998,2:605-610.
5 Bulas-Cruz J,Barroso J,Rafael A.Real-Time Number Plate Reading.4th IFAC Workshop on Algorithms and Architectures for Real-time Control.http://www.utad.pt.1997
6 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):574-575
7 王昱,赵正校,杨硕.基于直线边缘识别的图像区域定位方法.计算机工程.1999,25(9):61-62
8 Kamat V,Ganesan S.An efficient implementation of the Hough transform for detecting vehicle license plates using DSP'S.Real-Time Technology and Applications Symposium.May 1995:58-59
9 牛欣,沈兰荪.基于特征的车辆牌照定位算法.交通与计算机.2001,1(81):31-33
10 宴建华,赵正校,基于属性开运算的汽车牌照区域定位算法.红外与激光工程.2000,29(3):19-22
11 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.
12 白洪亮,娄正良,邹明福,等.复杂背景下基于形态学的车牌识别系统.公路交通科技.2004,21(10):117-120
13 Hsien-Huang P Wu,Hung-Hsiang Chen,Ruei-Jan Wu,et al.License Plate Extraction in Low Resolution Video.18th Intemational Conference on Pattern Recognition,ICPR.2006,1:824-827
14 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.中国图象图形学报.2005,5(5):411-415
15 许礼武,许伦辉,黄艳国.基于小波分解的车牌定位算法.计算机工程.2006, 32(21):191-193
16秦钟,馀建闽,史胜利,等.基于字符角点信息的车牌定位方法.华南理工大学学报自然科学版.2006,34(4):19-23.
17全书海,薛志华,王琴.基于投影图像分布特征的车牌定位算法研究与实现.武汉理工大学学报交通科学与工程版.2005,29(6):879-882
18刘文峰,吴学毅,刘长富.基于RGB色度空间的车牌定位及矫正算法.武汉大学学报信息科学版.2006,31(9):785-787
19贾小军,喻擎苍,谭召均.纹理谱描述子及其在车牌定位中的应用.计算机应用研究.200;7,24(3):215-217
20周泽华,潘保昌,郑胜林,等。基于多颜色模型的车牌定位方法.微计算机信息.2007,23(1):283-285
21 Nijhuis J A.Car License Plate Recognition with Neural Networks and Fuzzy Logic.Proceedings IEEE International Conference on Neural Networks.1995,5:2232-2236
22吴德会,王晓红,石俊,等.基于车牌特征颜色相似度的定位方法.公路交通科技.2005,22(1):135-139
23 Wenjing Jia,Huaifeng Zhang,Xiangjian He,et al.Mean shift for accurate license plate localization.IEEE Intelligent Conference on Transportation Systems,ITSC'05.2005,2005:566-571
24郭捷,施鹏飞.基于颜色和纹理分析的车牌定位方法.中国图象图形学报.2002,7(5):472-476
25张引,潘云鹤.彩色汽车图像牌照定位新方法.中国图像图形学报.2001,6(4):374-377
26李文举,梁德群,张旗,等.基于边缘颜色对的车牌定位新方法.计算机学报.2004,27(2):204-208
27刘军杰,张南.基于多层前馈神经网络的车牌提取方法.广东交通职业技术学院学报.2007,6(1):27-32
28 Ying-Nong Chen,Chin-Chuan Han,Cheng-Tzu Wang,et al.The Application of a Convolution Neural Network on Face and License Plate Detection.18th International Conference on Pattern Recognition,ICPR.2006,3:552-555
29 Viola P,Jones M.Rapid object detection using a boosted cascade of simple features.IEEE Cortference.on Computer Vision and Pattern Recognition.2001,1:511-518.
30灌石柱,殳伟群,壬令群.基于Adaboost的汽车牌照快速定位.计算机工程.2006,32(12):187-188,214
31 Zhang A,Jia W,He X,et al.Learning-based license plate detection using global and local features.18th International Conference on Pattern Recognition,ICPR.2006,2:1102-1105
32 Xiaowei Xu,Zhiyan Wang,Yanqing Zhang,et al.A method of multi-view vehicle license plates location based on rectangle features.The 8th International Conference on Signal Processing.2006,3:16-20
33 Dlagnekov L.License plate detection using Adaboost.La Jolla:Comput.Sci.Eng.Dept.,Univ.Califonia San Diego,Mar.2004.[Online].
34吴舟舟,李树广,基于分级边缘间距的实时车牌检测.中国图象图形学报.2007,12(2):315-321
35蒋治华,陈继荣,王卫.一种非传统车牌倾斜矫正方法.计算机应用研究.2006,3:175-177
36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
37 Bruce J,Balch T,Veloso M.Fast and Inexpensive Color Image Segmentation for Interactive Robots.IEEE International Conference on Intelligent Robots and Systems,2000.3:2061-2066
38边肇祺,张学工,等.模式识别.第2版.北京:清华大学出版社,2001
39刘琴.机器学习.武汉工程职业技术学院院报.2001,13(2):41-44
40董乐红,耿国华,高原.Boosting算法综述.计算机应用与软件.2006,23(8):27-29
41 Valiant L G.A Theory of Learnable.Communication of ACM,1984,27(11):1134-1142
42 Schapire R.E.,The Strength of Weak Learnability,Machine Learning,1990,5(2):197-227
43 Freund Y,Schapire R E.A Decision-theoretic Generalization of Online Learning and an Application to Boosting.Journal of Computer and System Sciences,1997,55(1):119-139
44沈利明.人脸跟踪算法及其在DAM6416平台上的实现.南京理工大学.2006
45 Ma Yong,Ding Xiaoqing.Real-Time Multi-View Face Detection and Pose Estimation Based on Cost-Sensitive AdaBoost.Ysinghua Science and Technologe,2005,10(2):152-157.
46 Fan W,Stolfo SJ,Zhang J,Chan PK.Adacost:misclassification cost-sensitive boosting.Proceedings of the Sixteenth International Conference on Machine Learning,1999:97-105
47王良红,冷建华.汽车倾斜牌照中字符的定位与提取.电讯技术.2003,4:59-62
48包明,路小波.基于Hough变换的车牌倾斜检测算法.交通与计算机.2004,22(2):57-60
49 Wen C Y,Yu C C,Hun Z D.A 3-D transformation to improve the legibility of license plate numbers.Journal of Forensic Sciences,2002,47(3):578-585
50李文举,梁德群,王新年.质量退化的车牌字符分割方法.计算机辅助设计与图形学学报.2004,16(5):697-700
51李文举,梁德群,崔连延.一种新的面向字符分割的车牌图像预处理方法.计算机应用研究.2004,7:258-260
52权伟,郑南宁,贾新春.复杂背景下的车辆牌照字符提取方法研究.信息与控制.2002,31(1):25-29
53崔江,王友仁.车牌自动识别方法中的关键技术研究.计算机测量与控制.2003,11(4):260-262
54 Hans A Hegt,De la Haye,Ron J,et al.A high performance license plate recognition system.IEEE International conference on systems,man,and cybernetics,1998,5:4357-4362
55彭启琮等.DSP集成开发环境——CCS及DSP/BIOS的原理与应用.北京:电子工业出版社,2004
56李方慧等.TMS320C6000系列DSPs原理与应用.第2版.北京:电子工业出版社,2002
57 Evaluation Module(EVM)for the TMS320DM642 Quick Start Installation Guide.S Spectrum Digital Inc.
58 TMS320C6000 Optimizing C/C++ Compiler User's Guide,Texas Instruments Incorporated,2001
59 TMS320C6000 Code Composer Studio User's Guide,Texas Instruments Incorporated,2000
60 TMS320 DSP/BIOS User's Guide.Texas Instruments Incorporated.2003
61 TMS320DM642 DSP Cache User's Guide.Texas Instruments Incorporated.May,2003
2 Barroso J,Dagless E L,Rafael A,et al.Number plate reading using computer vision.IEEE International Symposium on Industrial Electronics.1997,3:761-766
3 Parisi R,Di Claudio E D,Lucarelli G,et al.Car Plate Recognition by Neural Networks and Image Processing.IEEE Intemational Symposium on Circuits and Systems.1998,3:195-198
4 Coetzee C,Botha C,Weber D.PC Based Number Plate Recognition System.In Proc.IEEE International Symposium on Industrial Eleetronies.1998,2:605-610.
5 Bulas-Cruz J,Barroso J,Rafael A.Real-Time Number Plate Reading.4th IFAC Workshop on Algorithms and Architectures for Real-time Control.http://www.utad.pt.1997
6 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):574-575
7 王昱,赵正校,杨硕.基于直线边缘识别的图像区域定位方法.计算机工程.1999,25(9):61-62
8 Kamat V,Ganesan S.An efficient implementation of the Hough transform for detecting vehicle license plates using DSP'S.Real-Time Technology and Applications Symposium.May 1995:58-59
9 牛欣,沈兰荪.基于特征的车辆牌照定位算法.交通与计算机.2001,1(81):31-33
10 宴建华,赵正校,基于属性开运算的汽车牌照区域定位算法.红外与激光工程.2000,29(3):19-22
11 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.
12 白洪亮,娄正良,邹明福,等.复杂背景下基于形态学的车牌识别系统.公路交通科技.2004,21(10):117-120
13 Hsien-Huang P Wu,Hung-Hsiang Chen,Ruei-Jan Wu,et al.License Plate Extraction in Low Resolution Video.18th Intemational Conference on Pattern Recognition,ICPR.2006,1:824-827
14 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.中国图象图形学报.2005,5(5):411-415
15 许礼武,许伦辉,黄艳国.基于小波分解的车牌定位算法.计算机工程.2006, 32(21):191-193
16秦钟,馀建闽,史胜利,等.基于字符角点信息的车牌定位方法.华南理工大学学报自然科学版.2006,34(4):19-23.
17全书海,薛志华,王琴.基于投影图像分布特征的车牌定位算法研究与实现.武汉理工大学学报交通科学与工程版.2005,29(6):879-882
18刘文峰,吴学毅,刘长富.基于RGB色度空间的车牌定位及矫正算法.武汉大学学报信息科学版.2006,31(9):785-787
19贾小军,喻擎苍,谭召均.纹理谱描述子及其在车牌定位中的应用.计算机应用研究.200;7,24(3):215-217
20周泽华,潘保昌,郑胜林,等。基于多颜色模型的车牌定位方法.微计算机信息.2007,23(1):283-285
21 Nijhuis J A.Car License Plate Recognition with Neural Networks and Fuzzy Logic.Proceedings IEEE International Conference on Neural Networks.1995,5:2232-2236
22吴德会,王晓红,石俊,等.基于车牌特征颜色相似度的定位方法.公路交通科技.2005,22(1):135-139
23 Wenjing Jia,Huaifeng Zhang,Xiangjian He,et al.Mean shift for accurate license plate localization.IEEE Intelligent Conference on Transportation Systems,ITSC'05.2005,2005:566-571
24郭捷,施鹏飞.基于颜色和纹理分析的车牌定位方法.中国图象图形学报.2002,7(5):472-476
25张引,潘云鹤.彩色汽车图像牌照定位新方法.中国图像图形学报.2001,6(4):374-377
26李文举,梁德群,张旗,等.基于边缘颜色对的车牌定位新方法.计算机学报.2004,27(2):204-208
27刘军杰,张南.基于多层前馈神经网络的车牌提取方法.广东交通职业技术学院学报.2007,6(1):27-32
28 Ying-Nong Chen,Chin-Chuan Han,Cheng-Tzu Wang,et al.The Application of a Convolution Neural Network on Face and License Plate Detection.18th International Conference on Pattern Recognition,ICPR.2006,3:552-555
29 Viola P,Jones M.Rapid object detection using a boosted cascade of simple features.IEEE Cortference.on Computer Vision and Pattern Recognition.2001,1:511-518.
30灌石柱,殳伟群,壬令群.基于Adaboost的汽车牌照快速定位.计算机工程.2006,32(12):187-188,214
31 Zhang A,Jia W,He X,et al.Learning-based license plate detection using global and local features.18th International Conference on Pattern Recognition,ICPR.2006,2:1102-1105
32 Xiaowei Xu,Zhiyan Wang,Yanqing Zhang,et al.A method of multi-view vehicle license plates location based on rectangle features.The 8th International Conference on Signal Processing.2006,3:16-20
33 Dlagnekov L.License plate detection using Adaboost.La Jolla:Comput.Sci.Eng.Dept.,Univ.Califonia San Diego,Mar.2004.[Online].
34吴舟舟,李树广,基于分级边缘间距的实时车牌检测.中国图象图形学报.2007,12(2):315-321
35蒋治华,陈继荣,王卫.一种非传统车牌倾斜矫正方法.计算机应用研究.2006,3:175-177
36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
37 Bruce J,Balch T,Veloso M.Fast and Inexpensive Color Image Segmentation for Interactive Robots.IEEE International Conference on Intelligent Robots and Systems,2000.3:2061-2066
38边肇祺,张学工,等.模式识别.第2版.北京:清华大学出版社,2001
39刘琴.机器学习.武汉工程职业技术学院院报.2001,13(2):41-44
40董乐红,耿国华,高原.Boosting算法综述.计算机应用与软件.2006,23(8):27-29
41 Valiant L G.A Theory of Learnable.Communication of ACM,1984,27(11):1134-1142
42 Schapire R.E.,The Strength of Weak Learnability,Machine Learning,1990,5(2):197-227
43 Freund Y,Schapire R E.A Decision-theoretic Generalization of Online Learning and an Application to Boosting.Journal of Computer and System Sciences,1997,55(1):119-139
44沈利明.人脸跟踪算法及其在DAM6416平台上的实现.南京理工大学.2006
45 Ma Yong,Ding Xiaoqing.Real-Time Multi-View Face Detection and Pose Estimation Based on Cost-Sensitive AdaBoost.Ysinghua Science and Technologe,2005,10(2):152-157.
46 Fan W,Stolfo SJ,Zhang J,Chan PK.Adacost:misclassification cost-sensitive boosting.Proceedings of the Sixteenth International Conference on Machine Learning,1999:97-105
47王良红,冷建华.汽车倾斜牌照中字符的定位与提取.电讯技术.2003,4:59-62
48包明,路小波.基于Hough变换的车牌倾斜检测算法.交通与计算机.2004,22(2):57-60
49 Wen C Y,Yu C C,Hun Z D.A 3-D transformation to improve the legibility of license plate numbers.Journal of Forensic Sciences,2002,47(3):578-585
50李文举,梁德群,王新年.质量退化的车牌字符分割方法.计算机辅助设计与图形学学报.2004,16(5):697-700
51李文举,梁德群,崔连延.一种新的面向字符分割的车牌图像预处理方法.计算机应用研究.2004,7:258-260
52权伟,郑南宁,贾新春.复杂背景下的车辆牌照字符提取方法研究.信息与控制.2002,31(1):25-29
53崔江,王友仁.车牌自动识别方法中的关键技术研究.计算机测量与控制.2003,11(4):260-262
54 Hans A Hegt,De la Haye,Ron J,et al.A high performance license plate recognition system.IEEE International conference on systems,man,and cybernetics,1998,5:4357-4362
55彭启琮等.DSP集成开发环境——CCS及DSP/BIOS的原理与应用.北京:电子工业出版社,2004
56李方慧等.TMS320C6000系列DSPs原理与应用.第2版.北京:电子工业出版社,2002
57 Evaluation Module(EVM)for the TMS320DM642 Quick Start Installation Guide.S Spectrum Digital Inc.
58 TMS320C6000 Optimizing C/C++ Compiler User's Guide,Texas Instruments Incorporated,2001
59 TMS320C6000 Code Composer Studio User's Guide,Texas Instruments Incorporated,2000
60 TMS320 DSP/BIOS User's Guide.Texas Instruments Incorporated.2003
61 TMS320DM642 DSP Cache User's Guide.Texas Instruments Incorporated.May,2003