小波分析在图像分割和车牌识别中的应用研究
|
摘要
图像分割是计算机图像识别与理解中一个十分活跃的研究领域,是计算机图像理解方法实现的基础。主动轮廓模型是图像分割的一种重要的方法,能够准确地实现图像中目标对象的分割,因而成为众多研究者研究的热点问题。小波分析作为信号和图像处理的一种强有力的工具,其理论框架已基本形成。小波分析具有其他许多信号分析方法(比如Fourier变换、Gabor变换等)所不具备的一些优良特性。小波分析的多分辨特性能够提供图像在不同分辨率下不同通道上的图像特征。
本文在推广梯度矢量流的基础上,提出了基于二进小波变换的改进外力场的计算方法。通过综合利用图像小波分解的高频信息,迭代计算获得其外力场,使Snake在图像的多尺度空间中搜索目标轮廓。针对Snake模型对初始轮廓的依赖性问题,给出了在一维最大熵阈值分割后的图像上获取初始轮廓点的方法,减少了人工的干预,加快了Snake的收敛速度和效果。实验结果表明该模型能有效地排除噪声的干扰,搜索凹陷轮廓,而且对脆弱轮廓有很好的逼近能力。
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。近年来,对车牌识别系统中关键技术的研究已经成为智能交通领域的一个热点问题。
本文对车牌识别系统中的车牌定位,车牌图像倾斜校正以及字符分割等几大关键部分进行了比较深入,全面的论述。车牌定位是车牌识别的基础,针对复杂背景,复杂环境车牌定位难的问题,综合利用了车牌区域水平灰度跳变特征,自适应地实现图像的二值化;利用数学形态学对图像进行一系列形态运算,消除了大量无用信息和噪声;根据区域分析提取出候选车牌区域;最后结合先验知识,将车牌准确地定位出来。然后通过Harris算法得到车牌的角点信息,并在此基础上,根据Hough变换得到车牌的倾斜角度,实现车牌的倾斜校正,接着对车牌进行去除边框的处理,最后利用投影法实现车牌字符的切分。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和字符分割效果,具有一定的鲁棒性和实时性。
Image segmentation is thoroughly active research domain in the field of computer image recognition and comprehension. It’s the basic of image comprehension. Activate contour model is a important method of image segmentation, and can segment the object accurately, so it has already become an important research field. Wavelet transform is a powerful tool in the signal and image processing, and its fundamental theory has been formed. Wavelet transform has many properties which other signal analysis techniques (for example, Fourier analysis, Gabor Transform, etc) do not have. Because of the multiresolution property, the wavelet transform provides image’s characters on different channel and different scale.
This paper generalizes the gradient vector flow and then introduces an improved external force based on dyadic wavelet transform. The detail coefficients of image wavelet transform are used to derive the new external force, so the snake can track the object contours through the multi-scale space of an image. In additional, aimed at the snake’s dependence on initial contour points, one-dimensional maximum-entropy threshold algorithm is adopted to segment the image firstly and then obtain the initial contour points, which reduce the manual operation as well as improve the convergence of snake speed. Experimental results proved that the new model can greatly get rid of influence of noise and search concave contours. Furthermore, it is efficiently convergence to weak boundary.
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. In recent years, the study on the critical techniques for LPR has already become an important research field of scientific circles.
In this paper we discuss the main parts of LPR: license plate location, slant correction and character segmentation, and make a deep research on some important and key technology. Automatic vehicle location license plate is the basic of LPR. It’s difficult for realizing this technology in complex background and complex environment. According to a great deal of horizontal texture of vehicle image, some preprocesses such as adaptive binary conversion, a series of morphological operations which can greatly decrease the noise and useless information in vehicle image; then the method of region analysis can select some candidate regions; lastly using prior knowledge locate the license plate region accurately. Next step is the characters segmentation. We can get the corners of the license image using the Harris algorithm. At the same time, the slant angle of license plate is necessary to use the method combined with Hough transform and corner information. Then we excise the border of the characters and segment the characters. 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 and segment characters. It shows that the performance of the system is promising, robust and timely.
本文在推广梯度矢量流的基础上,提出了基于二进小波变换的改进外力场的计算方法。通过综合利用图像小波分解的高频信息,迭代计算获得其外力场,使Snake在图像的多尺度空间中搜索目标轮廓。针对Snake模型对初始轮廓的依赖性问题,给出了在一维最大熵阈值分割后的图像上获取初始轮廓点的方法,减少了人工的干预,加快了Snake的收敛速度和效果。实验结果表明该模型能有效地排除噪声的干扰,搜索凹陷轮廓,而且对脆弱轮廓有很好的逼近能力。
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。近年来,对车牌识别系统中关键技术的研究已经成为智能交通领域的一个热点问题。
本文对车牌识别系统中的车牌定位,车牌图像倾斜校正以及字符分割等几大关键部分进行了比较深入,全面的论述。车牌定位是车牌识别的基础,针对复杂背景,复杂环境车牌定位难的问题,综合利用了车牌区域水平灰度跳变特征,自适应地实现图像的二值化;利用数学形态学对图像进行一系列形态运算,消除了大量无用信息和噪声;根据区域分析提取出候选车牌区域;最后结合先验知识,将车牌准确地定位出来。然后通过Harris算法得到车牌的角点信息,并在此基础上,根据Hough变换得到车牌的倾斜角度,实现车牌的倾斜校正,接着对车牌进行去除边框的处理,最后利用投影法实现车牌字符的切分。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和字符分割效果,具有一定的鲁棒性和实时性。
Image segmentation is thoroughly active research domain in the field of computer image recognition and comprehension. It’s the basic of image comprehension. Activate contour model is a important method of image segmentation, and can segment the object accurately, so it has already become an important research field. Wavelet transform is a powerful tool in the signal and image processing, and its fundamental theory has been formed. Wavelet transform has many properties which other signal analysis techniques (for example, Fourier analysis, Gabor Transform, etc) do not have. Because of the multiresolution property, the wavelet transform provides image’s characters on different channel and different scale.
This paper generalizes the gradient vector flow and then introduces an improved external force based on dyadic wavelet transform. The detail coefficients of image wavelet transform are used to derive the new external force, so the snake can track the object contours through the multi-scale space of an image. In additional, aimed at the snake’s dependence on initial contour points, one-dimensional maximum-entropy threshold algorithm is adopted to segment the image firstly and then obtain the initial contour points, which reduce the manual operation as well as improve the convergence of snake speed. Experimental results proved that the new model can greatly get rid of influence of noise and search concave contours. Furthermore, it is efficiently convergence to weak boundary.
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. In recent years, the study on the critical techniques for LPR has already become an important research field of scientific circles.
In this paper we discuss the main parts of LPR: license plate location, slant correction and character segmentation, and make a deep research on some important and key technology. Automatic vehicle location license plate is the basic of LPR. It’s difficult for realizing this technology in complex background and complex environment. According to a great deal of horizontal texture of vehicle image, some preprocesses such as adaptive binary conversion, a series of morphological operations which can greatly decrease the noise and useless information in vehicle image; then the method of region analysis can select some candidate regions; lastly using prior knowledge locate the license plate region accurately. Next step is the characters segmentation. We can get the corners of the license image using the Harris algorithm. At the same time, the slant angle of license plate is necessary to use the method combined with Hough transform and corner information. Then we excise the border of the characters and segment the characters. 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 and segment characters. It shows that the performance of the system is promising, robust and timely.
引文
[1]章毓晋.中国图像工程2005[J].中国图像图形学报.11(5),2006:601-623.
[2]王慧燕,程翼宇.边缘检测中小波选取问题研究[J]电路与系统学报,11(23),2006:48-53.
[3]刘丽,王曙钊,李敬社.脊波变换的研究现状与应用前景[J]信息与电子工程,3(4),2005:315-319.
[4]唐晓初.小波分析及其应用[M].重庆:重庆大学出版社,2006.
[5]王志刚.多尺度多分辨率图像分割研究. [D]上海交通大学博士论文.2002.
[6] S.Mallat.A wavelet tour of signal processing[M],Second Edition机械工业出版社.2003.
[7] S.Mallat.Characterization of signals from multiscale edge[J].IEEE Transactions On Pattern Analysis And Machine Intelligence , 14(7),1992:710-732.
[8] YuanY.Tang. Characterization and Detection of Edges by Lipschitz Exponents and MASW Wavelet Transform.[C]14th International Conference on Pattern Recognition ,Australia, 1998, 1572—1574.
[9] L.FENG.C.Y.SUEN. Edge extraction of image by reconstruction using wavelet decomposition detail at different resolution levels.International journal of pattern and artificial intelligence[J] 14(16),2000:779-793.
[10] Duan Heric, Damjan Zazula. Combined edge detection using wavelet transform and signal registration Image and Vision Computing.25(5), 2007:652-662.
[11]刘静.基于小波变换的图像分割技术研究[D].山东大学硕士论文.2005.
[12] S.Mallat.Singularity detection and processing with wavelets.[J]IEEE Trans.Inform.Theory , 38(2),1992 :617-643.
[13] Kass M, Witkin A, Terzopoulos D. Snake: Active contourmodels[J]International Journal Computer Vision, 1(4), 1987:321-331.
[14] Xu C, Prince J L. Snakes, shape, and Gradient vector flow [J]. IEEE Transaction image Processing, March , 7(3), 1998:359-369.
[15] Li C, Liu J D, Fox M D. Segment of Edge Preserving Gradient Vector Flow: An Approach Toward Automatically Initializing and Splitting of Snake.[C] Proceeding of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2005:162-167.
[16]黄海赟,戚飞虎,赵雪春基于多尺度图像的主动轮廓线模型[J]计算机研究与发展.37(10), 2000:1240-1245.
[17]马波,张田文,李培华基于改进外部力的多尺度蛇算法[J].计算机学报26(5), 2003:563-568.
[18]张慧,刘伟军.一种基于小波的轮廓特征提取算法[J].中国图像图形学报,10(7),2005:828-833.
[19] Ming-Yu Shih.A wavelet based multiresolution edge detection and tracking.Image and vision computing[J] 23,2005:441-451.
[20] Wu H H, Liu J C,Chui C.A Wavelet-Frame Based Image Force Model for Active Contouring Algorithms.[J] IEEE Transaction on image processing, 9(11), 2000 :1983-1988.
[21] Sakalli M, Lam K M, Yan H. Shivering greedy snakes, gradient-guided in wavelet domain.[C] IEEE International Conference on Image Processing, Vol.2 , 1998:886-890.
[22]曾理,侯立华.基于改进Snake模型的图像分割方法[J]计算机仿真, 23(7), 2006 :180-182.
[23]李丽勤,高焕文,周兴祥.Snake模型初始轮廓选取的研究[J]计算机工程与应用.2004:43-45.
[24]王忠民,鲍旭东,舒华忠.一种新的活动轮廓模型算法[J]东南大学学报.33(4),2003:434-437.
[25]马社祥,刘贵忠,曾召华.由信号二进小波变换极值重构信号的样条插值算法[J]电子学报,29(1),2001:68-70.
[26]刘贵忠,张志明,冯牧,章珂.信号重构的小波极大模整形迭代算法[J]自然科学进展.10(7),2000:660-664.
[27]孙丰荣,翟广涛,李艳玲,刘泽,张梅,张运.由小波变换模极大值重构信号的快速算法[J]小型微型计算机系统,126(12),2005:2147-2149.
[28] Zeer Berman.The uniqueness question of discrete wavelet maxima representation[R]Tech.Rep. TR 91-84. Cent Univ.of Maryland, College Park, Apr, 1991.
[29] Ming Yu,Ying-Chun Guo,Zhi-Tao Xiao. A novel edge detection method and its application in ultrasound images. [C]Proceedings of the Third International Conference on Machine Learning and Cybernetics, 8,2004:3973-3976.
[30]黄山.车牌识别技术的研究和实现.四川大学博士论文.2005.
[31]袁宝民,于万波,魏小鹏.汽车牌照定位研究综述[J].大连大学学报,Vol.23,No.2,2002,7-12.
[32]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.Vol .6,No.4,2001,374-377.
[33] Xifan Shi,Weixhong Zhao,Yonghang Shen.Automatic License Recongnition System Based on Color Image Procsssing[J]. Computational Science and Its Applications.2005,5: 1159-1168.
[34] Min Hu, Xijian Ping, Yihong Ding. Automated cell nucleus segmentation using improved snake.[C] 2004 ICIP. 2737-3740.
[35] M. Stella Atkins, Blair T. Mackiewich .Fully Automatic Segmentation of the Brain in MRI[J] IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 17, NO.1, 1998,99-10.
[36]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]中国图像图形学报,Vol5,No.5,2000,411-415.
[37]许礼武,许伦辉,黄艳国.基于小波分解的车牌定位算法[J]计算机工程.32(21),2006,191-193.
[38]冯国进.车牌自动定位与模糊识别算法[J]光电子激光.Vol.14,No.7,2003,749-752.
[39] Xinlin Chen,Jie Yang,Jing Zhang. Automatic Detection and Recognition of Signs From Natural Scenes[J].IEEE TRANSACTION ON IMAGE PROCESSING.2004,1,13(1):87-99.
[40] Jiang Gao,Jie Yang.An Adaptive Algorithm for Text Detection From Natural Scenes[C]. Proceedings of the 2001 IEEE Conference on Computer Vision and Pattern Recognition. 2001,12:84-89.
[41]邱菊,马驷良,张忠波,基于小波变换的显微图像细胞边缘提取.[J]吉林大学学报.Vol43,No.3,2005,320-323.
[42]段瑞玲,李庆祥,李玉和.图像边缘检测方法研究综述[J]光学技术,Vol31,No3,2005,415-418.
[43]季虎,孙即祥,邵晓芳,毛玲.图像边缘提取方法及展望.[J]计算机工程与应用No.14,2004,70-73.
[44]张变莲,唐慧君,闫旻奇.一种复杂车辆图像中的多车牌定位方法[J].光子学报,2007,36(1):184-187.
[45]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.2004,2,33(1):29-33.
[46]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, 2004,27(1):130-135.
[47] Feng Yang,Zheng Ma,Mei Xie.A Novel Approach for License Plate Character Segmentation [C].2006 1ST IEEE Conference on Industrial Electronics and Applications. 2006,12:1-6.
[48]章毓晋.椭圆匹配法及其在序列细胞图像3-D批准中的应用[J].中国图象图形学报. 1997,9,2(8):574-578.
[49]李文举,梁德群等.质量退化的车牌字符分割方法[J].计算机辅助设计与图形学学报. 2004,5,16(5):697-700.
[50] HARRIS C, STEPHENS M.A Combined Corner and Edge Detector[C]Proceedings of Fourth Alvey Vision Conference. 1988, 147-151.
[51] NOBUYUKI OTSU. A Threshold Selection Method from Gray-Level Histograms[J]. IEEE TRANSACTIONS ON SYSTREMS, MAN, AND CYBERNETICS, 1979,1,9(1):62-66.
[52]后俊.车牌图像分割与智能字符识别方法的研究[D].合肥工业大学硕士论文.2004.
[53]是湘全,何苑凌,蔡孟波.遗传算法在车牌定位中的应用[J]公路交通科技,17(2),2000,33-36.
[54]刘九芬.小波理论及其在图象压缩和数字水印中的应用[D].浙江大学博士论文.2001.
[55]夏良正.数字图像处理(修订版)[M].南京:东南大学出版社,1999.
[2]王慧燕,程翼宇.边缘检测中小波选取问题研究[J]电路与系统学报,11(23),2006:48-53.
[3]刘丽,王曙钊,李敬社.脊波变换的研究现状与应用前景[J]信息与电子工程,3(4),2005:315-319.
[4]唐晓初.小波分析及其应用[M].重庆:重庆大学出版社,2006.
[5]王志刚.多尺度多分辨率图像分割研究. [D]上海交通大学博士论文.2002.
[6] S.Mallat.A wavelet tour of signal processing[M],Second Edition机械工业出版社.2003.
[7] S.Mallat.Characterization of signals from multiscale edge[J].IEEE Transactions On Pattern Analysis And Machine Intelligence , 14(7),1992:710-732.
[8] YuanY.Tang. Characterization and Detection of Edges by Lipschitz Exponents and MASW Wavelet Transform.[C]14th International Conference on Pattern Recognition ,Australia, 1998, 1572—1574.
[9] L.FENG.C.Y.SUEN. Edge extraction of image by reconstruction using wavelet decomposition detail at different resolution levels.International journal of pattern and artificial intelligence[J] 14(16),2000:779-793.
[10] Duan Heric, Damjan Zazula. Combined edge detection using wavelet transform and signal registration Image and Vision Computing.25(5), 2007:652-662.
[11]刘静.基于小波变换的图像分割技术研究[D].山东大学硕士论文.2005.
[12] S.Mallat.Singularity detection and processing with wavelets.[J]IEEE Trans.Inform.Theory , 38(2),1992 :617-643.
[13] Kass M, Witkin A, Terzopoulos D. Snake: Active contourmodels[J]International Journal Computer Vision, 1(4), 1987:321-331.
[14] Xu C, Prince J L. Snakes, shape, and Gradient vector flow [J]. IEEE Transaction image Processing, March , 7(3), 1998:359-369.
[15] Li C, Liu J D, Fox M D. Segment of Edge Preserving Gradient Vector Flow: An Approach Toward Automatically Initializing and Splitting of Snake.[C] Proceeding of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2005:162-167.
[16]黄海赟,戚飞虎,赵雪春基于多尺度图像的主动轮廓线模型[J]计算机研究与发展.37(10), 2000:1240-1245.
[17]马波,张田文,李培华基于改进外部力的多尺度蛇算法[J].计算机学报26(5), 2003:563-568.
[18]张慧,刘伟军.一种基于小波的轮廓特征提取算法[J].中国图像图形学报,10(7),2005:828-833.
[19] Ming-Yu Shih.A wavelet based multiresolution edge detection and tracking.Image and vision computing[J] 23,2005:441-451.
[20] Wu H H, Liu J C,Chui C.A Wavelet-Frame Based Image Force Model for Active Contouring Algorithms.[J] IEEE Transaction on image processing, 9(11), 2000 :1983-1988.
[21] Sakalli M, Lam K M, Yan H. Shivering greedy snakes, gradient-guided in wavelet domain.[C] IEEE International Conference on Image Processing, Vol.2 , 1998:886-890.
[22]曾理,侯立华.基于改进Snake模型的图像分割方法[J]计算机仿真, 23(7), 2006 :180-182.
[23]李丽勤,高焕文,周兴祥.Snake模型初始轮廓选取的研究[J]计算机工程与应用.2004:43-45.
[24]王忠民,鲍旭东,舒华忠.一种新的活动轮廓模型算法[J]东南大学学报.33(4),2003:434-437.
[25]马社祥,刘贵忠,曾召华.由信号二进小波变换极值重构信号的样条插值算法[J]电子学报,29(1),2001:68-70.
[26]刘贵忠,张志明,冯牧,章珂.信号重构的小波极大模整形迭代算法[J]自然科学进展.10(7),2000:660-664.
[27]孙丰荣,翟广涛,李艳玲,刘泽,张梅,张运.由小波变换模极大值重构信号的快速算法[J]小型微型计算机系统,126(12),2005:2147-2149.
[28] Zeer Berman.The uniqueness question of discrete wavelet maxima representation[R]Tech.Rep. TR 91-84. Cent Univ.of Maryland, College Park, Apr, 1991.
[29] Ming Yu,Ying-Chun Guo,Zhi-Tao Xiao. A novel edge detection method and its application in ultrasound images. [C]Proceedings of the Third International Conference on Machine Learning and Cybernetics, 8,2004:3973-3976.
[30]黄山.车牌识别技术的研究和实现.四川大学博士论文.2005.
[31]袁宝民,于万波,魏小鹏.汽车牌照定位研究综述[J].大连大学学报,Vol.23,No.2,2002,7-12.
[32]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.Vol .6,No.4,2001,374-377.
[33] Xifan Shi,Weixhong Zhao,Yonghang Shen.Automatic License Recongnition System Based on Color Image Procsssing[J]. Computational Science and Its Applications.2005,5: 1159-1168.
[34] Min Hu, Xijian Ping, Yihong Ding. Automated cell nucleus segmentation using improved snake.[C] 2004 ICIP. 2737-3740.
[35] M. Stella Atkins, Blair T. Mackiewich .Fully Automatic Segmentation of the Brain in MRI[J] IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 17, NO.1, 1998,99-10.
[36]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]中国图像图形学报,Vol5,No.5,2000,411-415.
[37]许礼武,许伦辉,黄艳国.基于小波分解的车牌定位算法[J]计算机工程.32(21),2006,191-193.
[38]冯国进.车牌自动定位与模糊识别算法[J]光电子激光.Vol.14,No.7,2003,749-752.
[39] Xinlin Chen,Jie Yang,Jing Zhang. Automatic Detection and Recognition of Signs From Natural Scenes[J].IEEE TRANSACTION ON IMAGE PROCESSING.2004,1,13(1):87-99.
[40] Jiang Gao,Jie Yang.An Adaptive Algorithm for Text Detection From Natural Scenes[C]. Proceedings of the 2001 IEEE Conference on Computer Vision and Pattern Recognition. 2001,12:84-89.
[41]邱菊,马驷良,张忠波,基于小波变换的显微图像细胞边缘提取.[J]吉林大学学报.Vol43,No.3,2005,320-323.
[42]段瑞玲,李庆祥,李玉和.图像边缘检测方法研究综述[J]光学技术,Vol31,No3,2005,415-418.
[43]季虎,孙即祥,邵晓芳,毛玲.图像边缘提取方法及展望.[J]计算机工程与应用No.14,2004,70-73.
[44]张变莲,唐慧君,闫旻奇.一种复杂车辆图像中的多车牌定位方法[J].光子学报,2007,36(1):184-187.
[45]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.2004,2,33(1):29-33.
[46]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, 2004,27(1):130-135.
[47] Feng Yang,Zheng Ma,Mei Xie.A Novel Approach for License Plate Character Segmentation [C].2006 1ST IEEE Conference on Industrial Electronics and Applications. 2006,12:1-6.
[48]章毓晋.椭圆匹配法及其在序列细胞图像3-D批准中的应用[J].中国图象图形学报. 1997,9,2(8):574-578.
[49]李文举,梁德群等.质量退化的车牌字符分割方法[J].计算机辅助设计与图形学学报. 2004,5,16(5):697-700.
[50] HARRIS C, STEPHENS M.A Combined Corner and Edge Detector[C]Proceedings of Fourth Alvey Vision Conference. 1988, 147-151.
[51] NOBUYUKI OTSU. A Threshold Selection Method from Gray-Level Histograms[J]. IEEE TRANSACTIONS ON SYSTREMS, MAN, AND CYBERNETICS, 1979,1,9(1):62-66.
[52]后俊.车牌图像分割与智能字符识别方法的研究[D].合肥工业大学硕士论文.2004.
[53]是湘全,何苑凌,蔡孟波.遗传算法在车牌定位中的应用[J]公路交通科技,17(2),2000,33-36.
[54]刘九芬.小波理论及其在图象压缩和数字水印中的应用[D].浙江大学博士论文.2001.
[55]夏良正.数字图像处理(修订版)[M].南京:东南大学出版社,1999.