车辆识别系统中的基础性算法研究
|
摘要
基于图像检测与处理技术的交通车辆识别系统是智能交通系统(Intelligent Transportation System, ITS)中提高交通运输效率、减少交通车辆犯罪等问题的基本手段之一。本文重点研究了基于图像检测与处理技术的交通车辆识别系统中核心技术之一的车辆牌照识别技术,对其涉及的车辆图像预处理、车辆牌照的定位、车辆牌照字符识别等关键技术进行了深入研究。
论文首次提出了一种基于自适应权值与非线性自适应预测相结合的椒盐噪声去除方法,它首先预判图像椒盐噪声的位置,随后对图像边界像素噪声、非图像边界像素噪声、图像的非噪声像素进行分别处理,从而实现车辆图像预处理中椒盐噪声的去除。
针对车辆图像预处理中高斯噪声的去除问题,借助原高阶偏微分方程模型的去噪思想,建立了一种基于极小曲面定理的高阶偏微分方程图像去噪模型,它在有效去除图像高斯噪声的同时保持了图像的细节,而且对于椒盐噪声也有抑制能力。
研究了模糊集合理论,提出了一种专门针对于车辆图像的Ⅱ型模糊集二值化算法,它在Ⅰ型模糊隶属度的基础上,建立Ⅱ型模糊集,利用遍历搜寻在最小模糊度意义下的最佳灰度值作为分块后的图像二值化阈值,完成了车辆牌照定位的基础工作。
分析了车辆牌照字符的纹理特征,设计了一种基于纹理分析的车辆牌照定位算法,它以二值图像为基础,利用改进的局域二值模式(Local Binary Pattern, LBP)算法进行图像的纹理分析,然后利用扫描法与投影法对车辆牌照进行定位。
基于Freeman链码知识,构建了一种用于车辆牌照字符识别的一维多重隐马尔可夫模型(Hidden Markov Model, HMM),它通过提取字符水平投影、垂直投影、目标点质心倾角投影及字符目标点距离投影的状态链码作为字符的特征,以此特征训练一维多重隐马尔可夫模型,从而实现车辆牌照字符识别。
在以上理论研究与数学推导基础上,通过计算机仿真验证了本文所推理论模型的可行性及有效性,为进一步的实际应用工作奠定了基础。
Traffic vehicle recognition system based on the image detecting technology and image processing technology is the basic means to improve traffic transportation efficiency and reduce traffic vehicle crimes in Intelligent Transportation System (ITS). The key research in this dissertation is one of the core technologies in vehicle recognition system, the vehicle license plate recognition technology. The vehicle image pre-processing, vehicle license plate location and vehicle license plate characters recognition in vehicle license plate recognition technology are deeply studied.
A novel image salt and pepper noise removal algorithm based on adaptive weight and nonlinear predication is proposed in this dissertation for the first time. It estimates the location of the noise in advance. Subsequently, the image boundary noise, the non image boundary noise and the non image noise are processed separately. So the salt and pepper noise removal are realized in vehicle image pre-processing.
Aiming at the Gaussian noise removal in vehicle image pre-processing, a novel high-order PDE noise removal model based on the Minimal Surface in differential geometry is designed by means of the original high-order PDE noise removal model theory. It removes image Gaussian noise and protected image details. Meanwhile, it has the image salt and pepper noise inhibition.
Fuzzy set theory is studied and a binarization algorithm based on the type II fuzzy set for car image is proposed. It establishes a type II fuzzy set based on the type I fuzzy membership and searches an optimal image gray level ergodicly as the image binarization threshold in the minimal fuzzy degree. So the basic task for vehicle license plate location is accomplished.
Vehicle license plate texture features are analyzed and a vehicle license plate location algorithm is designed based on the texture analysis. First it utilizes the Local Binary Pattern (LBP) algorithm to analyze the image texture based on the binarization image. Then scanning method and projection method are used to locate vehicle license plate.
A one-dimensional multiple Hidden Markov Model (HMM) for vehicle license plate character recognition is constructed based on the Freeman chain code. It extracts the character's horizontal projection chain code, vertical projection chain code, centroid inclination projection chain code of target pixel and distance projection chain code of target pixel as the character features. These features are used to train the character's one-dimensional multiple HMM and the vehicle license plate characters are recognized by this model.
Based on the theoretical study and mathematical derivation above, compute simulation verifies the theory model's feasibility and the efficiency. The foundation for the further practical application task is established.
论文首次提出了一种基于自适应权值与非线性自适应预测相结合的椒盐噪声去除方法,它首先预判图像椒盐噪声的位置,随后对图像边界像素噪声、非图像边界像素噪声、图像的非噪声像素进行分别处理,从而实现车辆图像预处理中椒盐噪声的去除。
针对车辆图像预处理中高斯噪声的去除问题,借助原高阶偏微分方程模型的去噪思想,建立了一种基于极小曲面定理的高阶偏微分方程图像去噪模型,它在有效去除图像高斯噪声的同时保持了图像的细节,而且对于椒盐噪声也有抑制能力。
研究了模糊集合理论,提出了一种专门针对于车辆图像的Ⅱ型模糊集二值化算法,它在Ⅰ型模糊隶属度的基础上,建立Ⅱ型模糊集,利用遍历搜寻在最小模糊度意义下的最佳灰度值作为分块后的图像二值化阈值,完成了车辆牌照定位的基础工作。
分析了车辆牌照字符的纹理特征,设计了一种基于纹理分析的车辆牌照定位算法,它以二值图像为基础,利用改进的局域二值模式(Local Binary Pattern, LBP)算法进行图像的纹理分析,然后利用扫描法与投影法对车辆牌照进行定位。
基于Freeman链码知识,构建了一种用于车辆牌照字符识别的一维多重隐马尔可夫模型(Hidden Markov Model, HMM),它通过提取字符水平投影、垂直投影、目标点质心倾角投影及字符目标点距离投影的状态链码作为字符的特征,以此特征训练一维多重隐马尔可夫模型,从而实现车辆牌照字符识别。
在以上理论研究与数学推导基础上,通过计算机仿真验证了本文所推理论模型的可行性及有效性,为进一步的实际应用工作奠定了基础。
Traffic vehicle recognition system based on the image detecting technology and image processing technology is the basic means to improve traffic transportation efficiency and reduce traffic vehicle crimes in Intelligent Transportation System (ITS). The key research in this dissertation is one of the core technologies in vehicle recognition system, the vehicle license plate recognition technology. The vehicle image pre-processing, vehicle license plate location and vehicle license plate characters recognition in vehicle license plate recognition technology are deeply studied.
A novel image salt and pepper noise removal algorithm based on adaptive weight and nonlinear predication is proposed in this dissertation for the first time. It estimates the location of the noise in advance. Subsequently, the image boundary noise, the non image boundary noise and the non image noise are processed separately. So the salt and pepper noise removal are realized in vehicle image pre-processing.
Aiming at the Gaussian noise removal in vehicle image pre-processing, a novel high-order PDE noise removal model based on the Minimal Surface in differential geometry is designed by means of the original high-order PDE noise removal model theory. It removes image Gaussian noise and protected image details. Meanwhile, it has the image salt and pepper noise inhibition.
Fuzzy set theory is studied and a binarization algorithm based on the type II fuzzy set for car image is proposed. It establishes a type II fuzzy set based on the type I fuzzy membership and searches an optimal image gray level ergodicly as the image binarization threshold in the minimal fuzzy degree. So the basic task for vehicle license plate location is accomplished.
Vehicle license plate texture features are analyzed and a vehicle license plate location algorithm is designed based on the texture analysis. First it utilizes the Local Binary Pattern (LBP) algorithm to analyze the image texture based on the binarization image. Then scanning method and projection method are used to locate vehicle license plate.
A one-dimensional multiple Hidden Markov Model (HMM) for vehicle license plate character recognition is constructed based on the Freeman chain code. It extracts the character's horizontal projection chain code, vertical projection chain code, centroid inclination projection chain code of target pixel and distance projection chain code of target pixel as the character features. These features are used to train the character's one-dimensional multiple HMM and the vehicle license plate characters are recognized by this model.
Based on the theoretical study and mathematical derivation above, compute simulation verifies the theory model's feasibility and the efficiency. The foundation for the further practical application task is established.
引文
[1]Ron Zahavi. Enterprise Application Integration with CORBA [M]. New York:John Wiley & Sons,1999.
[2]纪天明,贺跃,于同,王少军.智能车辆导航系统中的实时道路检测[J].计算机应用,2005,25(12)：228-232.
[3]陈燕龙.车辆牌照检测关键算法研究[J].电脑与电信,2007,10：56-58.
[4]Brownrigg D. R. K. The weighted median filter[J]. Communications of the ACM, 1984,27(8):807-818.
[5]Sun-Jea Ko, Yong Hong Lee. Center weighted median filters and their applications to image enhancement [J]. IEEE Trans on Circuits and Syst,1991,38(9):984-993.
[6]Tong Sun, Neuvo Y. Detail-preserving median based filters in image processing [J]. Pattern Recognition Letters,1994,15(4):341-347.
[7]H.Hwang, R.A.Haddad. Adaptive median filters:New algorithms and results [J]. IEEE Trans on Image Processing,1995,4(4):499-502.
[8]Nallaperumal, K. Varghese, J. Saudia, S. Selvakumar, R. K. Krishnaveni, K. Vinsley, S. S. Selective switching median filter for the removal of salt & pepper impulse noise [C].2006 IFIP International Conference on Wireless and Optical Communications Networks. Bangalore, India,2006.
[9]Hirokazu Yagou, Yutaka Ohtakey, Alexander Belyaevz. Mesh Smoothing via Mean and Median Filtering Applied to Face Normals[C]. Geometric Modeling and Processing IEEE, Japan,2002.
[10]胡浩,王明照,杨杰.自适应模糊加权均值滤波器[J].系统工程与电子技术,2002,24(2)：5-17.
[11]周登文,申晓留.邻域小波系数自适应得图像去噪[J].中国图象图形学报,2008,13(11)：2112-2116.
[12]Perona P, Malik. Scale-space and edge detection using anisotropic diffusion[J]. IEEE Trans Pattern Analysis and Machine Intelligence,1990, 12(7):629-639.
[13]Yu-Li You, Kaveh M. Fourth-Order partial differential equations for noise removal [J]. IEEE Trans on Image Processing,2000,9(10):1723—1730.
[14]赵志刚,万娇娜,管聪慧.基于小波包变换与自适应阈值的图像去噪[J].中国图象图形学报,2007,12(6)：977-980.
[15]张树波,赖剑煌.车牌定位和分割的一种综合方法[J].中山大学学报(自然科学版),2004,43(2)：126-132.
[16]Shyang-Lih Chang, Li-Shien Chen, Yun-Chung Chung, Sei-Wan Chen. Automatic License Plate Recognition [J]. IEEE Trans. on Intelligent Transportation Systems,2004,5(1):42-53.
[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]张引,潘云鹤.彩色汽车图象牌照定位新方法[J].中国图象图形学报,2001,6(4)：374-377.
[19]郭捷,施鹏飞.基于彩色和纹理分析的车牌定位方法[J].中国图象图形学报,2002,7(5)：472-476.
[20]王枚,房培玉,王国宏.基于彩色分量垂直边缘检测的车牌定位新方法[J].计算机工程与应用,2007,43(12)：177-180.
[21]张引.基于空间分布的最大类间方差牌照图像二值化算法[J].浙江大学学报,2001,35(3)：272-275.
[22]叶晨洲,廖金周.一种基于纹理的牌照图像二值化方法[J].微型电脑应用,1999,15(6)：28-29.
[23]权炜,郑南宁,贾新春.复杂背景下的车辆牌照字符提取方法研究[J].信息与控制,2002,31(1)：25-29.
[24]骆雪超,刘杜雄,冯云庆,申伯华.一种基于车牌特征信息的车牌识别方法[J].华南理工大学学报,2003,31(4)：70-73.
[25]Barroso J, Bulas-cluz J, Dagless E L. Real-Time Number Plate Reading[C].4th IFAC Workshop on Algorithms and Architectures for Real-Time Control, Vilamoura, Portugal,1997.
[26]冯晓毅,周凤岐.一种基于点密度和分类约束规则的图像边缘提取方法[J].弹箭与制导学报,1999,19(1)：1-4.
[27]Trier 0. D, Jain A. K. Goal-directed evaluation of binarization methods [J].IEEE Trans on Pattern Analysis and Machine Intelligence,1995,17(12): 1191-1201.
[28]许礼武,许伦辉,黄艳国.基于小波分解的车牌定位方法[J].计算机工程,2006,32(21)：191-193.
[29]黄卫,路小波,余彦翔,凌小静.基于纹理及小波分析的车牌定位方法[J].中国工程科学,2004,6(3)：19-24.
[30]杨俊,戚飞虎.一种基于形状和纹理特征的车牌定位方法[J].计算机工程,2006,32(2)：170-172.
[31]路小波,张光华.基于二值图像的车牌精确定位方法[J].东南大学学报,2005,35(6)：972-974.
[32]刘广起,郑晓势,张晓波.基于图像纹理特征提取的车牌定位算法[J].中国图象图形学报,2005,10(11)：1419-1422.
[33]穆长江,苑玮琦.基于纹理特征的车牌定位方法[J].控制工程,2004,11(6)：574-576.
[34]郑伯川,崔屏,张征.一种基于粗细定位相结合的车牌定位方法[J].重庆邮电大学学报,2007,19(2)：225-227.
[35]童剑军,邹明福.车牌字符的一种精判别识别方法[J].计算机工程与应用,2005,41(2)：223-226.
[36]马俊莉,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.
[37]刘维一,盛益强,乔辉,方志良.特征点匹配法实现汽车牌照的快速识别[J].光电子.激光,2002,13(3)：274-276.
[38]Lee seong-Whan, Kim Young Joon. Direct Extraction of topographic feature for
gray scale character recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(7):724-729.
[39]黄戈祥,陈继荣,胡应添.基于ART2神经网络的车牌字符识别[J].计算机应用,2004,24(z2)：309-312.
[40]王敏,黄心汉.一种模板匹配和神经网络的车牌字符识别方法[J.华中科技大学学报,2001,29(3)：48-50.
[41]赵先军.基于神经网络的车牌识别技术研究[学位论文].西安：西安电子科技大学,2005.
[42]高珊,刘万春,朱玉文.基于SVM的车牌字符分割和识别方法[J].微电子学与计算机,2005,22(6)：34-36.
[43]任俊.基于支撑矢量机的图像分类、车牌识别及其嵌入式应用研究[学位论文].浙江：浙江大学,2005.
[44]陈来荣,冀荣华,徐宇.基于支持向量基的车牌字符识别[J].公路交通科技,2006,23(5)：124-127.
[45]葛宝臻,陈希明,张以漠.基于光电混合联合变换相关的汽车牌照识别方法的实验研究[J].光电子.激光,1999,10(3)：248-251.
[46]曲新亮,魏守水,杜洋.一种基于小波变换的车牌字符识别方法[J].山东建筑工程学院学报,2004,19(3)：54-56.
[47]潘翔,叶修梓,张三元.基于小波的车牌汉字特征提取[J].中国图象图形学报,2003,8(10)：1218-1222.
[48]陈友仁,赵正校.基于隐马尔可夫模型的车牌自动识别技术[J].红外与激光工程,2001,30(2)：101-107.
[49]曹刚.运动车辆识别技术研究[学位论文].四川：四川大学,2004.
[50]张毓晋.图像工程[M].北京：清华大学出版社,2007.
[51]W. B. Horng, M. T. Wu, J. H. Fann. Efficiently locating vehicle license plates based on vertical line detection [C].12th IPPR Conference on Computer Vision Graphics and Image Processing, Taiwan,1999.
[52]丁润涛.非线性滤波器的新进展[J].通信学报,1988,19(2)：37-44.
[53]Rafael C Gonzalez, Richard E Woods.数字图像处理[M].阮秋琦,阮宇智等译.北京：电子工业出版社,2003.
[54]王保平,李宗领.一种新型的去除图像中强脉冲噪声滤波器[J].计算机工程与应用,2003,39(20)：51-52.
[55]董继扬,张军英.一种简单的椒盐噪声滤除算法[J].计算机工程与应用,2003,39(20)：27-28.
[56]梁雯,刘松林.图像中心加权中值滤波的改进与应用[J].中国图像图形学报,1997,(2)：629-633.
[57]胡海涛.图像分析方法在指纹识别中的应用[学位论文].广州：中山大学,1991.
[58]曲大健.数学形态学在指纹图像处理中的应用及指纹比对算法[学位论文].北京：清华大学,1989.
[59]冯星奎.指纹图像的获取及其预处理、后处理[学位论文].西安：第二炮兵工程学院,1998.
[60]冯星奎,肖兴明,尹洪君.方向加权中值滤波算法[J].中国图象图形学报,2000,5(7)： 609-611.
[61]邢藏菊,王守觉等.一种基于极值中值的新型滤波算法[J].中国图象图形学报,2001,6(6)：533-536.
[62]Han W Y, Lin J C. Minimum-maximum exclusive mean filter to remove impulse noise from highly corrupted images [J]. IEEE Electronics Letters,1997,33 (2):124-125.
[63]宋宇,李满天等.基于相似度函数的图像椒盐噪声自适应滤除算法[J].自动化学报,2007,33(5)：474-479.
[64]Wang Z, Zhang D. Progressive switching median filter for the removal of impulse noise from highly corrupted images [J]. IEEE Trans on Circuits and systems-II: Analog and Digital Signal Processing,1999,46(1):78-80.
[65]孙即详等.图像压缩与投影重建[M].北京：科学出版社,2005.
[66]刘培,王建英,尹忠科.利用线性预测去除图像椒盐噪声[J].铁道学报,2007,29(6)：48-51.
[67]王益艳等.基于2级噪声检测的改进中值滤波算法[J].计算机工程与设计,2008,29(11)：2841-2843.
[68]高鑫,刘来福.基于PDE模型的图像处理方法[J].数学的实践与认识,2001,31(2)：206-210.
[69]F. Guichard, J.-M. Morel. Geometric partial differential equations and iterative filtering[C]. Proceedings of the fourth international symposium on Mathematical morphology and its applications to image and signal processing, Amsterdam, Netherlands,1998.
[70]Joachim Weickert. Coherence-Enhancing Diffusion Filtering [J]. International Journal of Computer Vision.1999,31(2):111-127.
[71]陈维桓.极小曲面[M].湖南：湖南教育出版社,1993.
[72]Kenneth R C. Digital image processing [M]. Engle wood Cliffs:Prentice-Hall International, Inc.,1988.
[73]耿修瑞,黎锁平.用于图像去噪的一个四阶偏微分方程[J].甘肃工业大学学报,2002,28(4)：119-121.
[74]Li Rui, Zhang Yujin. A hybrid filter for the cancellation of mixed Gaussian noise and impulse noise[C]. Proceedings of the 2003 Joint Conference of the Fourth International Conference on Information, Communications and Signal Processing,2003 and the Fourth Pacific Rim Conference on Multimedia, Beijing, 2003.
[75]H. R. Tizhoosh, Fuzzy Image Processing, Springer, Heidelberg, Germany,1998.
[76]王积分,张新荣.计算机图像识别[M].北京：中国铁道出版社,1988.
[77]W. Niblack, Wayne. An Introduction to Digital Image Processing [M]. Englewood Cliffs, N. J.:Prentice Hall,1986.
[78]J. Bernsen. Dynamic Thresholding of gray level Image [C]. Proceedings of the 8th International Conference on Pattern Recognition, Los Angeles,1986.
[79]L A Zadeh. The concept of a linguistic variable and its application to approximate reasoning-1[J]. Information Sciences,1975,8:199-249.
[80]Liang Q, Mendel J M. MPEG VBR video traffic modeling and classification and classification using fuzzy techniques [J]. IEEE Transactions on Fuzzy Systtem. 2001,9(1):183-193.
[81]Liang Q, Mendel J M. Overcoming time-varying co-channel interference using type-2 fuzzy adaptive filter [J]. IEEE Trans. on Circuits Syst.-Ⅱ:Analog and Digital Signal Processing,2000,47:1419-1428.
[82]Zarandi M H F, Neshar E, Turksen I B, Rezaee B. A Type-2 Fuzzy Model for Stock Market Analysis[C]. IEEE international conference on Fuzzy Systems 2007, London,2007.
[83]Innocent P R, John R I, Belton I, Finlay D. Type 2 fuzzy representation of lung scans to predict pulmonary emboli[C]. IFSA World Congress and the 20th NAFIPS International Conference, Vancouver, BC, Canada,2001.
[84]H. R. Tizhoosh. Image thresholding using type Ⅱ fuzzy sets [J], Pattern Recognition,2005,38:2363-2372.
[85]L.K.Huang, M.J.Wang. Image thresholding by minimizing the measure of fuzziness [J]. Pattern Recognition,1995,28(1):41-51.
[86]A. Kaufmann. Introduction to the Theory of Fuzzy Subsets—Fundamental Theoretical Elements [M]. Academic Press, New York,1975.
[87]W. A. Yasnoff, J. K. Mui, J. W. Bacus. Error measures for scenesegmentation [J]. Pattern Recognition,1977,9:217-231.
[88]张亶,陈刚.基于偏微分方程的图像处理[M].北京：高等教育出版社.2004.
[89]J. L. Chen, A. Kundu. Rotation and gray scale transform invariant texture identification using wavelet decomposition and hidden Markovmodel[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence,1994,16:208-214.
[90]Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray scale and rotation invariant texture analysis with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(7):971-987.
[91]Hongliang Jin, Qingshan Liu, Hanqing Lu, Xiaofeng Tong. Face detection using improved LBP under Bayesian Framework[C]. Proceedings of the Third International Conference on Image and Graphics, Hong Kong,2004.
[92]黄志斌.面向车辆牌照的L快速二值化算法[J].华侨大学学报,2002,23(4)：427-430.
[93]H. A. Hegt, R. J. Haye, N. A. Khan. A high performance license plate recognition system[C]. IEEE Conference on Systems, Man, and Cybernetics,1998.
[94]黄山.车牌识别技术的研究和实现[学位论文].四川：四川大学,2005.
[95]宋加涛,刘济林.车辆牌照上英文和数字字符的结构特征分析与提取[J].中国图象图形学报,2002,7(9)：945-949.
[96]值俊文,杨晓敏,刘大宇等.一种基于径向神经网络的车牌字符识别方法[J].计算机应用,2005,25(12)：311-313.
[97]袁贞明,陈焕宇.一种识别汽车牌照的字符分割技术[J].计算机时代,2004,(8)：5-6.
[98]宋建才.汽车牌照识别技术研究[J].工业控制计算机,2004,17(4)：44-45.
[99]刘斌,王普等.货车车号识别的研究与应用[J].电视技术,1999,(7)：38-40.
[100]高庆吉.对粘连和缺损数字串分割的研究[J].模式识别与人工智能,2000,13(1)：100-102.
[101]李文举.智能交通中图像处理技术应用的研究[学位论文].辽宁：大连海事大学,2004.
[102]王强.车牌照识别关键技术研究[学位论文].陕西：西北大学,2007.
[103]L R Rabiner, Juang B. An introduction to hidden markov models[J]. IEEE ASSP Magazine, 1986,3(1):4-16.
[104]L R Rabiner. A tutorial on hidden markov models and select applications in speech recognition [C]. Proceeding of IEEE,1989.
[105]R Durbin, S Eddy, A Krogh, G Mitchison.生物序列分析,蛋白质和核酸的概率论模型[M].北京：清华大学出版社,2002.
[106]张引,潘云鹤.工程图纸自动输入字符识别的二维隐性马尔可夫模型方法[J].计算机辅助设计与图形学学报,1999,11(5)：403-406.
[107]冯兵,丁晓表,吴佑寿.HMM方法识别脱机手写汉字[J].模式识别与人工智能,2002,15(1)：84-88.
[108]刘云中,林亚平,陈治平等.基于隐马尔可夫模型的文本信息抽取[J].系统仿真学报,2004,16(3)：507-510.
[109]林亚平,刘云中,周顺先等.基于最大熵的隐马尔可夫模型文本信息抽取[J].电子学报,2005,33(2)：236-240.
[110]周昌乐.手写汉字的机器识别[M].北京：科学出版社,1997.
[111]Shi Da-ming, Gunn S R, Damper R I. Handwritten Chinese radical recognition using nonlinear active shape models [J]. IEEE Trans.on Pattern Analysis and Machine Intelligence,2003,25(2):277-280.
[112]Sameh M.Awaidah, Sabri A.Mahmoud. A multiple feature/resolution scheme to Arabic (Indian) numerals recognition using hidden Markov models [J]. Signal Processing,2009, 89(6):1176-1184.
[113]刘刚,张洪刚,郭军.用于脱机手写体识别的隐马尔可夫模型[J].计算机研究与发展,2003,40(8)：1252-1256.
[114]Jie Li, Jiaxin Wang, Yannon Zhao, Zehong Yang. A new approach for off-line handwritten Chinese character recognition using self-adaptive HMM [C]. The Fifth World Congress on Intelligent Control and Automation, Zhejiang,2004.
[115]K H Fielding, D W Ruck. Spatio-temporal pattern recognition rising hidden markov models [J]. IEEE Trans. on Aespace and Electronic Systems,1995,31(4):1292-1300.
[116]R Paul, Runkle, K Priya.. Hidden markov models for Multiaspect Target Classification [J]. IEEE Trans.on Signal Processing,1999,47(7):2035-3040.
[117]Jinhai Cai, Zhiqiang Liu. Integeration fo structural and statistical information for unconstrained handwritten numeral recognition[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence,1999,21(3):263-270.
[118]Y.He, Kunden.2-D shape classification using hidden markov model[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence,1991,13(11):1172-1184.
[119]Simmon Gunter, Horst Bunke. HMM-based handwritten word recognition:on the optimization of the number of states,training iterations and Gaussian components [J]. Pattern Recognition,2004,37(10):2069-2079.
[120]Peter Jancovic, Munewer Kokuer. Incorporating the voicing information into HMM-based automatic speech recognition in noisy environments [J]. Speech communication,2009, 51(5):438-451.
[121]Alexander Seward. A fast HMM match algorithm for very large vocabulary speech recognition[J]. Speech Communication,2004,42(2):191-206.
[122]Min-Sub Kim, Daijin Kim, Sang-Youn Lee. Face recognition using the embedded HMM with second-order block-specific observations[J]. Pattern Recognition,2003,36(11): 2723-2735.
[123]Paulo R.Cavalin, Robert Sabourin, Ching Y.Suen. Evaluation of incremental learning algorithms for HMM in the recognition of alphanumeric characters[J]. Pattern Recognition, 2009,42(12):3241-3253.
[124]M. S.Khorsheed. Offline recognition of omnifont Arabic text using the HMM ToolKit (HTK)[J]. Pattern Recognition Letters,2007,28(12):1563-1571.
[125]Y.Zhu, L.C.De Silva, C.C.Ko. Using moment invariants and HMM in facial expression recognition[J]. Pattern Recognition Letters,2002,23(1-3):83-91.
[126]Alessandro Vinciarelli, Samy Bengio. Writer adaptation techniques in HMM based Off-line Cursive Script Recognition[J]. Pattern Recognition Letters,2002,23(8):905-916.
[127]Jianying Hu, Sok Gek Lim, Michael K.Brown. Writer independent on-line handwriting recognition using an HMM approach[J]. Pattern Recognition,2000,33(1):133-147.
[128]Q.H.He, S.Kwong, K.F.Man, K.S.Tang. An improved maximum model distance approach for HMM-based speech recognition systems[J]. Pattern Recognition,2000,33(10): 1749-1758.
[129]C.Pisarn, T.Theeramunkong. An HMM-based method for Thai spelling speech recognition[J]. Computers & Mathematics with Applications,2007,185(2):759-768.
[130]Jen-Tzung Chien. Online hierarchical transformation of hidden Markov models for speech recognition[J]. IEEE Transactions on Speech and Audio Processing,1999,7(6):656-667.
[131]Yuting Qi, Paisley J.W, Carin.L. Music Analysis Using Hidden Markov Mixture Models[J]. IEEE Transactions on Signal Processing,2007,55(11):5209-5224.
[132]Jen-Tzung Chien, Chin-Pin Liao. Maximum Confidence Hidden Markov Modeling for Face Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008, 30(4):606-616.
[133]Singh GB, Haiping Song. Using Hidden Markov Models in Vehicular Crash Detection[J]. IEEE Transactions on Vehicular Technology,,2009,58(3):1119-1128.
[134]A P Dempster, N M Laird, D B Rubin. Maximum likelihood from incomplete data via the EM algorithm [J]. Journal of the Royal Statistical Society,1977,39(1):1-38.
[135]P Baldi, Y Chauvin. Smooth on-line learning algorithms for hidden Markov models[J]. Neural Computation.1994,6(2):307-318.
[136]王萍,刘恒,狄光敏.基于简约码特征树的字母和数字识别[J].天津大学学报,2008,41(6)：667-672.
[137]宋克欧,黄凤岗,兰晓亭.二值图像目标质心快速搜索跟踪算法[J].模式识别与人工智能,1988,11(2)：161-168.
[138]Zhang J, Tan T. Affine invariant classification and retrieval of texture images [J]. Pattern Recognition,2003,36(3):657-664.
[139]王庆.脱机手写体汉字识别方法研究[学位论文].陕西：西北工业大学,2000.
[2]纪天明,贺跃,于同,王少军.智能车辆导航系统中的实时道路检测[J].计算机应用,2005,25(12)：228-232.
[3]陈燕龙.车辆牌照检测关键算法研究[J].电脑与电信,2007,10：56-58.
[4]Brownrigg D. R. K. The weighted median filter[J]. Communications of the ACM, 1984,27(8):807-818.
[5]Sun-Jea Ko, Yong Hong Lee. Center weighted median filters and their applications to image enhancement [J]. IEEE Trans on Circuits and Syst,1991,38(9):984-993.
[6]Tong Sun, Neuvo Y. Detail-preserving median based filters in image processing [J]. Pattern Recognition Letters,1994,15(4):341-347.
[7]H.Hwang, R.A.Haddad. Adaptive median filters:New algorithms and results [J]. IEEE Trans on Image Processing,1995,4(4):499-502.
[8]Nallaperumal, K. Varghese, J. Saudia, S. Selvakumar, R. K. Krishnaveni, K. Vinsley, S. S. Selective switching median filter for the removal of salt & pepper impulse noise [C].2006 IFIP International Conference on Wireless and Optical Communications Networks. Bangalore, India,2006.
[9]Hirokazu Yagou, Yutaka Ohtakey, Alexander Belyaevz. Mesh Smoothing via Mean and Median Filtering Applied to Face Normals[C]. Geometric Modeling and Processing IEEE, Japan,2002.
[10]胡浩,王明照,杨杰.自适应模糊加权均值滤波器[J].系统工程与电子技术,2002,24(2)：5-17.
[11]周登文,申晓留.邻域小波系数自适应得图像去噪[J].中国图象图形学报,2008,13(11)：2112-2116.
[12]Perona P, Malik. Scale-space and edge detection using anisotropic diffusion[J]. IEEE Trans Pattern Analysis and Machine Intelligence,1990, 12(7):629-639.
[13]Yu-Li You, Kaveh M. Fourth-Order partial differential equations for noise removal [J]. IEEE Trans on Image Processing,2000,9(10):1723—1730.
[14]赵志刚,万娇娜,管聪慧.基于小波包变换与自适应阈值的图像去噪[J].中国图象图形学报,2007,12(6)：977-980.
[15]张树波,赖剑煌.车牌定位和分割的一种综合方法[J].中山大学学报(自然科学版),2004,43(2)：126-132.
[16]Shyang-Lih Chang, Li-Shien Chen, Yun-Chung Chung, Sei-Wan Chen. Automatic License Plate Recognition [J]. IEEE Trans. on Intelligent Transportation Systems,2004,5(1):42-53.
[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]张引,潘云鹤.彩色汽车图象牌照定位新方法[J].中国图象图形学报,2001,6(4)：374-377.
[19]郭捷,施鹏飞.基于彩色和纹理分析的车牌定位方法[J].中国图象图形学报,2002,7(5)：472-476.
[20]王枚,房培玉,王国宏.基于彩色分量垂直边缘检测的车牌定位新方法[J].计算机工程与应用,2007,43(12)：177-180.
[21]张引.基于空间分布的最大类间方差牌照图像二值化算法[J].浙江大学学报,2001,35(3)：272-275.
[22]叶晨洲,廖金周.一种基于纹理的牌照图像二值化方法[J].微型电脑应用,1999,15(6)：28-29.
[23]权炜,郑南宁,贾新春.复杂背景下的车辆牌照字符提取方法研究[J].信息与控制,2002,31(1)：25-29.
[24]骆雪超,刘杜雄,冯云庆,申伯华.一种基于车牌特征信息的车牌识别方法[J].华南理工大学学报,2003,31(4)：70-73.
[25]Barroso J, Bulas-cluz J, Dagless E L. Real-Time Number Plate Reading[C].4th IFAC Workshop on Algorithms and Architectures for Real-Time Control, Vilamoura, Portugal,1997.
[26]冯晓毅,周凤岐.一种基于点密度和分类约束规则的图像边缘提取方法[J].弹箭与制导学报,1999,19(1)：1-4.
[27]Trier 0. D, Jain A. K. Goal-directed evaluation of binarization methods [J].IEEE Trans on Pattern Analysis and Machine Intelligence,1995,17(12): 1191-1201.
[28]许礼武,许伦辉,黄艳国.基于小波分解的车牌定位方法[J].计算机工程,2006,32(21)：191-193.
[29]黄卫,路小波,余彦翔,凌小静.基于纹理及小波分析的车牌定位方法[J].中国工程科学,2004,6(3)：19-24.
[30]杨俊,戚飞虎.一种基于形状和纹理特征的车牌定位方法[J].计算机工程,2006,32(2)：170-172.
[31]路小波,张光华.基于二值图像的车牌精确定位方法[J].东南大学学报,2005,35(6)：972-974.
[32]刘广起,郑晓势,张晓波.基于图像纹理特征提取的车牌定位算法[J].中国图象图形学报,2005,10(11)：1419-1422.
[33]穆长江,苑玮琦.基于纹理特征的车牌定位方法[J].控制工程,2004,11(6)：574-576.
[34]郑伯川,崔屏,张征.一种基于粗细定位相结合的车牌定位方法[J].重庆邮电大学学报,2007,19(2)：225-227.
[35]童剑军,邹明福.车牌字符的一种精判别识别方法[J].计算机工程与应用,2005,41(2)：223-226.
[36]马俊莉,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.
[37]刘维一,盛益强,乔辉,方志良.特征点匹配法实现汽车牌照的快速识别[J].光电子.激光,2002,13(3)：274-276.
[38]Lee seong-Whan, Kim Young Joon. Direct Extraction of topographic feature for
gray scale character recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(7):724-729.
[39]黄戈祥,陈继荣,胡应添.基于ART2神经网络的车牌字符识别[J].计算机应用,2004,24(z2)：309-312.
[40]王敏,黄心汉.一种模板匹配和神经网络的车牌字符识别方法[J.华中科技大学学报,2001,29(3)：48-50.
[41]赵先军.基于神经网络的车牌识别技术研究[学位论文].西安：西安电子科技大学,2005.
[42]高珊,刘万春,朱玉文.基于SVM的车牌字符分割和识别方法[J].微电子学与计算机,2005,22(6)：34-36.
[43]任俊.基于支撑矢量机的图像分类、车牌识别及其嵌入式应用研究[学位论文].浙江：浙江大学,2005.
[44]陈来荣,冀荣华,徐宇.基于支持向量基的车牌字符识别[J].公路交通科技,2006,23(5)：124-127.
[45]葛宝臻,陈希明,张以漠.基于光电混合联合变换相关的汽车牌照识别方法的实验研究[J].光电子.激光,1999,10(3)：248-251.
[46]曲新亮,魏守水,杜洋.一种基于小波变换的车牌字符识别方法[J].山东建筑工程学院学报,2004,19(3)：54-56.
[47]潘翔,叶修梓,张三元.基于小波的车牌汉字特征提取[J].中国图象图形学报,2003,8(10)：1218-1222.
[48]陈友仁,赵正校.基于隐马尔可夫模型的车牌自动识别技术[J].红外与激光工程,2001,30(2)：101-107.
[49]曹刚.运动车辆识别技术研究[学位论文].四川：四川大学,2004.
[50]张毓晋.图像工程[M].北京：清华大学出版社,2007.
[51]W. B. Horng, M. T. Wu, J. H. Fann. Efficiently locating vehicle license plates based on vertical line detection [C].12th IPPR Conference on Computer Vision Graphics and Image Processing, Taiwan,1999.
[52]丁润涛.非线性滤波器的新进展[J].通信学报,1988,19(2)：37-44.
[53]Rafael C Gonzalez, Richard E Woods.数字图像处理[M].阮秋琦,阮宇智等译.北京：电子工业出版社,2003.
[54]王保平,李宗领.一种新型的去除图像中强脉冲噪声滤波器[J].计算机工程与应用,2003,39(20)：51-52.
[55]董继扬,张军英.一种简单的椒盐噪声滤除算法[J].计算机工程与应用,2003,39(20)：27-28.
[56]梁雯,刘松林.图像中心加权中值滤波的改进与应用[J].中国图像图形学报,1997,(2)：629-633.
[57]胡海涛.图像分析方法在指纹识别中的应用[学位论文].广州：中山大学,1991.
[58]曲大健.数学形态学在指纹图像处理中的应用及指纹比对算法[学位论文].北京：清华大学,1989.
[59]冯星奎.指纹图像的获取及其预处理、后处理[学位论文].西安：第二炮兵工程学院,1998.
[60]冯星奎,肖兴明,尹洪君.方向加权中值滤波算法[J].中国图象图形学报,2000,5(7)： 609-611.
[61]邢藏菊,王守觉等.一种基于极值中值的新型滤波算法[J].中国图象图形学报,2001,6(6)：533-536.
[62]Han W Y, Lin J C. Minimum-maximum exclusive mean filter to remove impulse noise from highly corrupted images [J]. IEEE Electronics Letters,1997,33 (2):124-125.
[63]宋宇,李满天等.基于相似度函数的图像椒盐噪声自适应滤除算法[J].自动化学报,2007,33(5)：474-479.
[64]Wang Z, Zhang D. Progressive switching median filter for the removal of impulse noise from highly corrupted images [J]. IEEE Trans on Circuits and systems-II: Analog and Digital Signal Processing,1999,46(1):78-80.
[65]孙即详等.图像压缩与投影重建[M].北京：科学出版社,2005.
[66]刘培,王建英,尹忠科.利用线性预测去除图像椒盐噪声[J].铁道学报,2007,29(6)：48-51.
[67]王益艳等.基于2级噪声检测的改进中值滤波算法[J].计算机工程与设计,2008,29(11)：2841-2843.
[68]高鑫,刘来福.基于PDE模型的图像处理方法[J].数学的实践与认识,2001,31(2)：206-210.
[69]F. Guichard, J.-M. Morel. Geometric partial differential equations and iterative filtering[C]. Proceedings of the fourth international symposium on Mathematical morphology and its applications to image and signal processing, Amsterdam, Netherlands,1998.
[70]Joachim Weickert. Coherence-Enhancing Diffusion Filtering [J]. International Journal of Computer Vision.1999,31(2):111-127.
[71]陈维桓.极小曲面[M].湖南：湖南教育出版社,1993.
[72]Kenneth R C. Digital image processing [M]. Engle wood Cliffs:Prentice-Hall International, Inc.,1988.
[73]耿修瑞,黎锁平.用于图像去噪的一个四阶偏微分方程[J].甘肃工业大学学报,2002,28(4)：119-121.
[74]Li Rui, Zhang Yujin. A hybrid filter for the cancellation of mixed Gaussian noise and impulse noise[C]. Proceedings of the 2003 Joint Conference of the Fourth International Conference on Information, Communications and Signal Processing,2003 and the Fourth Pacific Rim Conference on Multimedia, Beijing, 2003.
[75]H. R. Tizhoosh, Fuzzy Image Processing, Springer, Heidelberg, Germany,1998.
[76]王积分,张新荣.计算机图像识别[M].北京：中国铁道出版社,1988.
[77]W. Niblack, Wayne. An Introduction to Digital Image Processing [M]. Englewood Cliffs, N. J.:Prentice Hall,1986.
[78]J. Bernsen. Dynamic Thresholding of gray level Image [C]. Proceedings of the 8th International Conference on Pattern Recognition, Los Angeles,1986.
[79]L A Zadeh. The concept of a linguistic variable and its application to approximate reasoning-1[J]. Information Sciences,1975,8:199-249.
[80]Liang Q, Mendel J M. MPEG VBR video traffic modeling and classification and classification using fuzzy techniques [J]. IEEE Transactions on Fuzzy Systtem. 2001,9(1):183-193.
[81]Liang Q, Mendel J M. Overcoming time-varying co-channel interference using type-2 fuzzy adaptive filter [J]. IEEE Trans. on Circuits Syst.-Ⅱ:Analog and Digital Signal Processing,2000,47:1419-1428.
[82]Zarandi M H F, Neshar E, Turksen I B, Rezaee B. A Type-2 Fuzzy Model for Stock Market Analysis[C]. IEEE international conference on Fuzzy Systems 2007, London,2007.
[83]Innocent P R, John R I, Belton I, Finlay D. Type 2 fuzzy representation of lung scans to predict pulmonary emboli[C]. IFSA World Congress and the 20th NAFIPS International Conference, Vancouver, BC, Canada,2001.
[84]H. R. Tizhoosh. Image thresholding using type Ⅱ fuzzy sets [J], Pattern Recognition,2005,38:2363-2372.
[85]L.K.Huang, M.J.Wang. Image thresholding by minimizing the measure of fuzziness [J]. Pattern Recognition,1995,28(1):41-51.
[86]A. Kaufmann. Introduction to the Theory of Fuzzy Subsets—Fundamental Theoretical Elements [M]. Academic Press, New York,1975.
[87]W. A. Yasnoff, J. K. Mui, J. W. Bacus. Error measures for scenesegmentation [J]. Pattern Recognition,1977,9:217-231.
[88]张亶,陈刚.基于偏微分方程的图像处理[M].北京：高等教育出版社.2004.
[89]J. L. Chen, A. Kundu. Rotation and gray scale transform invariant texture identification using wavelet decomposition and hidden Markovmodel[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence,1994,16:208-214.
[90]Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray scale and rotation invariant texture analysis with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(7):971-987.
[91]Hongliang Jin, Qingshan Liu, Hanqing Lu, Xiaofeng Tong. Face detection using improved LBP under Bayesian Framework[C]. Proceedings of the Third International Conference on Image and Graphics, Hong Kong,2004.
[92]黄志斌.面向车辆牌照的L快速二值化算法[J].华侨大学学报,2002,23(4)：427-430.
[93]H. A. Hegt, R. J. Haye, N. A. Khan. A high performance license plate recognition system[C]. IEEE Conference on Systems, Man, and Cybernetics,1998.
[94]黄山.车牌识别技术的研究和实现[学位论文].四川：四川大学,2005.
[95]宋加涛,刘济林.车辆牌照上英文和数字字符的结构特征分析与提取[J].中国图象图形学报,2002,7(9)：945-949.
[96]值俊文,杨晓敏,刘大宇等.一种基于径向神经网络的车牌字符识别方法[J].计算机应用,2005,25(12)：311-313.
[97]袁贞明,陈焕宇.一种识别汽车牌照的字符分割技术[J].计算机时代,2004,(8)：5-6.
[98]宋建才.汽车牌照识别技术研究[J].工业控制计算机,2004,17(4)：44-45.
[99]刘斌,王普等.货车车号识别的研究与应用[J].电视技术,1999,(7)：38-40.
[100]高庆吉.对粘连和缺损数字串分割的研究[J].模式识别与人工智能,2000,13(1)：100-102.
[101]李文举.智能交通中图像处理技术应用的研究[学位论文].辽宁：大连海事大学,2004.
[102]王强.车牌照识别关键技术研究[学位论文].陕西：西北大学,2007.
[103]L R Rabiner, Juang B. An introduction to hidden markov models[J]. IEEE ASSP Magazine, 1986,3(1):4-16.
[104]L R Rabiner. A tutorial on hidden markov models and select applications in speech recognition [C]. Proceeding of IEEE,1989.
[105]R Durbin, S Eddy, A Krogh, G Mitchison.生物序列分析,蛋白质和核酸的概率论模型[M].北京：清华大学出版社,2002.
[106]张引,潘云鹤.工程图纸自动输入字符识别的二维隐性马尔可夫模型方法[J].计算机辅助设计与图形学学报,1999,11(5)：403-406.
[107]冯兵,丁晓表,吴佑寿.HMM方法识别脱机手写汉字[J].模式识别与人工智能,2002,15(1)：84-88.
[108]刘云中,林亚平,陈治平等.基于隐马尔可夫模型的文本信息抽取[J].系统仿真学报,2004,16(3)：507-510.
[109]林亚平,刘云中,周顺先等.基于最大熵的隐马尔可夫模型文本信息抽取[J].电子学报,2005,33(2)：236-240.
[110]周昌乐.手写汉字的机器识别[M].北京：科学出版社,1997.
[111]Shi Da-ming, Gunn S R, Damper R I. Handwritten Chinese radical recognition using nonlinear active shape models [J]. IEEE Trans.on Pattern Analysis and Machine Intelligence,2003,25(2):277-280.
[112]Sameh M.Awaidah, Sabri A.Mahmoud. A multiple feature/resolution scheme to Arabic (Indian) numerals recognition using hidden Markov models [J]. Signal Processing,2009, 89(6):1176-1184.
[113]刘刚,张洪刚,郭军.用于脱机手写体识别的隐马尔可夫模型[J].计算机研究与发展,2003,40(8)：1252-1256.
[114]Jie Li, Jiaxin Wang, Yannon Zhao, Zehong Yang. A new approach for off-line handwritten Chinese character recognition using self-adaptive HMM [C]. The Fifth World Congress on Intelligent Control and Automation, Zhejiang,2004.
[115]K H Fielding, D W Ruck. Spatio-temporal pattern recognition rising hidden markov models [J]. IEEE Trans. on Aespace and Electronic Systems,1995,31(4):1292-1300.
[116]R Paul, Runkle, K Priya.. Hidden markov models for Multiaspect Target Classification [J]. IEEE Trans.on Signal Processing,1999,47(7):2035-3040.
[117]Jinhai Cai, Zhiqiang Liu. Integeration fo structural and statistical information for unconstrained handwritten numeral recognition[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence,1999,21(3):263-270.
[118]Y.He, Kunden.2-D shape classification using hidden markov model[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence,1991,13(11):1172-1184.
[119]Simmon Gunter, Horst Bunke. HMM-based handwritten word recognition:on the optimization of the number of states,training iterations and Gaussian components [J]. Pattern Recognition,2004,37(10):2069-2079.
[120]Peter Jancovic, Munewer Kokuer. Incorporating the voicing information into HMM-based automatic speech recognition in noisy environments [J]. Speech communication,2009, 51(5):438-451.
[121]Alexander Seward. A fast HMM match algorithm for very large vocabulary speech recognition[J]. Speech Communication,2004,42(2):191-206.
[122]Min-Sub Kim, Daijin Kim, Sang-Youn Lee. Face recognition using the embedded HMM with second-order block-specific observations[J]. Pattern Recognition,2003,36(11): 2723-2735.
[123]Paulo R.Cavalin, Robert Sabourin, Ching Y.Suen. Evaluation of incremental learning algorithms for HMM in the recognition of alphanumeric characters[J]. Pattern Recognition, 2009,42(12):3241-3253.
[124]M. S.Khorsheed. Offline recognition of omnifont Arabic text using the HMM ToolKit (HTK)[J]. Pattern Recognition Letters,2007,28(12):1563-1571.
[125]Y.Zhu, L.C.De Silva, C.C.Ko. Using moment invariants and HMM in facial expression recognition[J]. Pattern Recognition Letters,2002,23(1-3):83-91.
[126]Alessandro Vinciarelli, Samy Bengio. Writer adaptation techniques in HMM based Off-line Cursive Script Recognition[J]. Pattern Recognition Letters,2002,23(8):905-916.
[127]Jianying Hu, Sok Gek Lim, Michael K.Brown. Writer independent on-line handwriting recognition using an HMM approach[J]. Pattern Recognition,2000,33(1):133-147.
[128]Q.H.He, S.Kwong, K.F.Man, K.S.Tang. An improved maximum model distance approach for HMM-based speech recognition systems[J]. Pattern Recognition,2000,33(10): 1749-1758.
[129]C.Pisarn, T.Theeramunkong. An HMM-based method for Thai spelling speech recognition[J]. Computers & Mathematics with Applications,2007,185(2):759-768.
[130]Jen-Tzung Chien. Online hierarchical transformation of hidden Markov models for speech recognition[J]. IEEE Transactions on Speech and Audio Processing,1999,7(6):656-667.
[131]Yuting Qi, Paisley J.W, Carin.L. Music Analysis Using Hidden Markov Mixture Models[J]. IEEE Transactions on Signal Processing,2007,55(11):5209-5224.
[132]Jen-Tzung Chien, Chin-Pin Liao. Maximum Confidence Hidden Markov Modeling for Face Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008, 30(4):606-616.
[133]Singh GB, Haiping Song. Using Hidden Markov Models in Vehicular Crash Detection[J]. IEEE Transactions on Vehicular Technology,,2009,58(3):1119-1128.
[134]A P Dempster, N M Laird, D B Rubin. Maximum likelihood from incomplete data via the EM algorithm [J]. Journal of the Royal Statistical Society,1977,39(1):1-38.
[135]P Baldi, Y Chauvin. Smooth on-line learning algorithms for hidden Markov models[J]. Neural Computation.1994,6(2):307-318.
[136]王萍,刘恒,狄光敏.基于简约码特征树的字母和数字识别[J].天津大学学报,2008,41(6)：667-672.
[137]宋克欧,黄凤岗,兰晓亭.二值图像目标质心快速搜索跟踪算法[J].模式识别与人工智能,1988,11(2)：161-168.
[138]Zhang J, Tan T. Affine invariant classification and retrieval of texture images [J]. Pattern Recognition,2003,36(3):657-664.
[139]王庆.脱机手写体汉字识别方法研究[学位论文].陕西：西北工业大学,2000.