印刷体文字识别的研究
|
摘要
笔划代表着汉字的内部特征,笔划穿越次数是对笔划进行全穿越,反应了汉字的整体特征,全穿越在粗分时区分汉字的能力不是太强,增加了二级识别的工作量。本文除了提取笔划全穿越外还提取笔划半穿越,并把半穿越的次数进行重新组合形成新的特征值。把全穿越和半穿越结合起来作为汉字的特征值,对汉字进行粗分,粗分不能区分的汉字,采用四个角的能量值密度特征对汉字进行细分。实验结果表明了该方法的有效性。与单独使用全穿透方法相比,本文提出的方法在粗分时区分汉字的能力增强,减少了二级识别的工作量。本文还对印刷体数字进行了研究,从数字的结构形状着手,通过分析印刷体数字的形状,提出了一种基于结构形状的印刷体数字识别方法。该方法不用对字符图像进行复杂的细化处理,减少了因细化带来的误差问题,因而识别速度非常快,实验证明了该方法的有效性。
Stokes represent internal character of Chinese Character, The previous method of traversing times of strokes is full-breakthrough to stroke, but this method is not effective for some Chinese Characters. This paper introduces half-breakthrough of strokes, and makes traversing times combine newly then obtains a new feature. It is used to be the first recognition with the combination of full-breakthrough and half-breakthrough. If it can not be recognized then make the second recognition with energy-density. This method does not need to complex thin to the character picture, reducing the erroneous question which is brought by thinning。The result shows this method is effective. The effect of the new method has obvious progress compared with the full-breakthrough only in first recognition, decreasing workload of the second recognition. A method of printing digital has been proposed that based on the structure shape, through analyzing the structure shape of the printing digital. This method does not need to complex thin to the character picture, reducing the erroneous question which is brought by thinning, so the recognition speed is quickly. The result shows this method is effective.
Stokes represent internal character of Chinese Character, The previous method of traversing times of strokes is full-breakthrough to stroke, but this method is not effective for some Chinese Characters. This paper introduces half-breakthrough of strokes, and makes traversing times combine newly then obtains a new feature. It is used to be the first recognition with the combination of full-breakthrough and half-breakthrough. If it can not be recognized then make the second recognition with energy-density. This method does not need to complex thin to the character picture, reducing the erroneous question which is brought by thinning。The result shows this method is effective. The effect of the new method has obvious progress compared with the full-breakthrough only in first recognition, decreasing workload of the second recognition. A method of printing digital has been proposed that based on the structure shape, through analyzing the structure shape of the printing digital. This method does not need to complex thin to the character picture, reducing the erroneous question which is brought by thinning, so the recognition speed is quickly. The result shows this method is effective.
引文
[1]Y.X.Gu,Q.R.Wang and C.Y.Suen,Application of multilayer decision tree in computer recognition of Chinese characters,IEEE Trans, Pattern Anal, Mach. Intell, 1983,5.83-89
[2]Y.Liu and T.Kasvand,A new approach to machine recognition of Chinese characters,Proc.7th Int.Conf.Pattern Recognition Montreal. Montreal Canada. PP. 1984.381-384
[3]张忻中.汉字识别技术.清华大学出版社.1992
[4]朱小燕,史一凡,马少平.手写体字符识别研究.模式识别与人工智能,2000 Vo113, No,2,p174-180
[5]张中.汉字识别技术综述.语言文学应用.1997 (2)
[6]张德喜.手写体机器识别技术的现状分析.许昌师专学报.1999(3)
[7]K.Mori and LMasuda, Advances in Recognition of Chinese characters, Proc. Of 5th Inten.conf. on Pattern Recognition, 1980.692-720
[8] Casey R Nagy G.Recognition of printed Chinese character.IEEE Trans.On Elec Comput,1966,15.91-101
[9]金连文.手写体汉字识别的研究:[博士学位论文].广州:华南理工大学.1996
[10]张忻中.汉字自动识别研究综述.中文信息.1984:11-14
[11]王芳.模式识别技术及其在文字识别领域的应用与研究:[硕士学位论文].西安.西北工业大学.2002
[12]吴全,朱兆达.图像处理中灰度级阈值选取方法 30 年(1962-1992)的进展. (数据采集与处理,1993,8:193-201
[13]PKsahoo, S.Soltani, A.K.C. Wong, and YC.Chen, A Survey of thresholding Techniques, Computer Graphics Vision and Image Processing 1988,41. 233-260
[14]瞿洋,杨利平.Hough 变换 OCR 图像倾斜矫正方法,中国图像图形学报,2001,6(2):178-181
[15]赵楠楠,欧阳鑫玉,吴庆洪.线性回归与 Hough 变换在计算图像倾斜角度中的应用.鞍山科技大学学报.2003,26(6):457-460
[16]X.YJiang, H.Bunke, D.Widmer-Kljajo, Skew detection of document images by focused nearest-neighbor clustering, 1999 IEEE International Conference on Document Analysis and Recognition, 1999.629-632
[17]N.Liolios, N.Fakotakis, G.Kokkinakis, Improved document skew detection based on text line connected-component clustering , 2001 IEEE International Conference on Image Processing, 2001, Vo1.1.1098-1101
[18]J.F.Cullen, K.Ejiri, Weak model-dependent page segmentation and skew correction for processing document images, 1993 IEEE International Conference on Document Analysis and Recognition, 1993.757-760
[19]H.Y.Ma, Z.W.Yu, An enhanced skew angle estimation technique for binary document images, 1999 IEEE International Conference on Document Analysis and Recognition, 1999.165-168
[20]R.Safabakhsh, S.Khadivi, Document skew detection using minimum-area bounding rectangle, 2000 IEEE International Conference on Information Technology, 2000.253-258
[21]王正群.手写体汉字识别研究:[博士学位论文].南京.南京理工大学.2001
[22]丁慧东.脱机手写体汉字识别研究:[硕士学位论文].长春.东北师范大学.2005
[23]冯志敏.基于结构特征的手写体汉字识别研究:[硕士学位论文].昆明.云南师范大学.2006
[24]周光博.一种印刷体汉字特征点提取的新方法: [硕士学位论文].上海.华东师范大学.2006
[25]Romero R D,Touretzky D ,Thibadeaun R H.Optical Chinese character recognition using probabilistic neural networks.Pattern Recognition,1997,30 (8).1279-129.
[26]Cheng Lin Liu,In Jung Kim,Jin H Kim.Model-based stroke extraction and matching for handwritten Chinese character recognition. Pattern Recognition,2001(34).2339-2352
[27]Y Mizukami.A handwritten Chinese character recognition system using Hierarchical displacement extraction based on directional features.Pattern Recognition Letters,1998,(19).595-604
[28]赵明.手写印刷体汉字识别方法综述.计算机研究与发展.1993 ,Vo1.30,No.4,59-64
[29]高彦宇,杨扬.脱机手写体汉字识别研究综述.计算机工程.2004 (7)
[30]路浩如,杨源远.手写体汉字识别问题综述.计算机应用与软件,Vol.l1
[31]丁晓青.汉字识别研究回顾.电子学报.2002 (9)
[32]汪芳,康慕宁等.印刷体汉字识别技术.情报技术.2004 (2)
[33] 柳 回 春 , 马 树 元 , 吴 平 东 等 . 手 写 体 数 字 识 别 技 术 的 研 究 . 计 算 机 工程.2003,29(4):24-25,61
[34]居琰,汪同庆,刘建胜,王贵新,彭健. 基于集成 RBF 神经网络的小类别手写体汉字识别系统.计算机工程与应用.2002.38(23):100-102,158
[35]张世辉,孔令富.汉字识别及现状分析.燕山大学学报. 2003,27(4):367-369
[36]梁涌.印刷体汉字识别系统的研究与实现:[硕士学位论文].西安.西北工业大学. 2006.3
[37]张 鸽,陈书 开.基于SVM的手写 体 阿 拉 伯 数 字 识 别 .军 民 两 用 技 术 与 产品.2005(9):41-43
[38]吴炜,杨晓敏,刘大宇等.一种基于模糊模板匹配的车牌汉字识别方法.微型机与应用.2005,24(11):57-59
[39]崔国伟,舒文豪,李仲荣.多字体多字号印刷汉字识别方法的研究.中文信息学报.1990,4(3):57-62
[40]Nadal C,Legault R,Suen C Y.Complementary algorithms for the recognition of totally unconstrained handwritten numerals[C].In:Proc of the l0th Int Conf Pattern Recognition,1990-06.434-449
[41]Cheng H Yan ,Recogninition of handwritten digits based on contour Information.Pattern Recognition,1998;31(3).235-255
[42]F.H. Cheng and W.H.Hsu,Research on Chinese OCR in Taiwan,Int ,J, Pattern Recognition Artificia1 Intell 5,(1&2).139-164(1991)
[43] C Downton,S Impedovo et a1.Progress in Handwriting Recongnition[M] World Scicentific Publishing Co Pte Ltd, 1997
[44]J Zhou,Q Gan,C Y Suen . A high performance hand-printed numeral recognition system with verification module[C].In:Proc of 4 ICDAR Ulm, Germany, 1997.294-297
[45]王 勇,吴立德.图像恢复和边缘提取的后验均值方法.电子学报.1994,22(2): 70-75
[46]武强,童学锋.季隽.基于人工神经网络的数字字符识别.计算机工程.2003. 8
[47]边肇棋.模式识别.北京:清华大学出版社.1998
[48]Martin T.Hagan, Howard B.Demuth, Mark H.Beale.神经网络设计.北京:机械工业出版社
[49]万红梅,金连文,尹俊勋等.结合距离分类器的神经网络手写体识别.计算机工程与应用,2004. 11
[50]Vapnik V.,The Nature of Statistical leanning Thory.New York: Springer- Verlag,1995
[51]Dong Jian-xiong, Krzyzak A, Suen Ching Y .An improved handwritten Chinese character recognition system using support vector machine .Pattern Recognition Letter.2005,26.1849-1856
[52]石繁槐,童学锋.SVM 在小字符集脱机手写体汉字识别中的应用研究.计算机工程, 2002.6
[53]乔海晔,肖南峰.基于视觉的文字识别系统的设计与实现.交通与计算机,2005(5)
[54]张重阳,娄震,杨静字.基于轮廓和统计特征的手写体数字识别.计算机工程与应用,2004,40(9):83-84,89
[55]娄震,胡钟山,杨静字.基于轮廓分段特征的手写体阿拉伯数字识别.计算机学报, 1999,22(10):1065-1073
[56]郭戈,闫继宏,蒋红梅等.基于结构特征的汉字识别.甘肃工业大学学报. 29(1):81-85
[2]Y.Liu and T.Kasvand,A new approach to machine recognition of Chinese characters,Proc.7th Int.Conf.Pattern Recognition Montreal. Montreal Canada. PP. 1984.381-384
[3]张忻中.汉字识别技术.清华大学出版社.1992
[4]朱小燕,史一凡,马少平.手写体字符识别研究.模式识别与人工智能,2000 Vo113, No,2,p174-180
[5]张中.汉字识别技术综述.语言文学应用.1997 (2)
[6]张德喜.手写体机器识别技术的现状分析.许昌师专学报.1999(3)
[7]K.Mori and LMasuda, Advances in Recognition of Chinese characters, Proc. Of 5th Inten.conf. on Pattern Recognition, 1980.692-720
[8] Casey R Nagy G.Recognition of printed Chinese character.IEEE Trans.On Elec Comput,1966,15.91-101
[9]金连文.手写体汉字识别的研究:[博士学位论文].广州:华南理工大学.1996
[10]张忻中.汉字自动识别研究综述.中文信息.1984:11-14
[11]王芳.模式识别技术及其在文字识别领域的应用与研究:[硕士学位论文].西安.西北工业大学.2002
[12]吴全,朱兆达.图像处理中灰度级阈值选取方法 30 年(1962-1992)的进展. (数据采集与处理,1993,8:193-201
[13]PKsahoo, S.Soltani, A.K.C. Wong, and YC.Chen, A Survey of thresholding Techniques, Computer Graphics Vision and Image Processing 1988,41. 233-260
[14]瞿洋,杨利平.Hough 变换 OCR 图像倾斜矫正方法,中国图像图形学报,2001,6(2):178-181
[15]赵楠楠,欧阳鑫玉,吴庆洪.线性回归与 Hough 变换在计算图像倾斜角度中的应用.鞍山科技大学学报.2003,26(6):457-460
[16]X.YJiang, H.Bunke, D.Widmer-Kljajo, Skew detection of document images by focused nearest-neighbor clustering, 1999 IEEE International Conference on Document Analysis and Recognition, 1999.629-632
[17]N.Liolios, N.Fakotakis, G.Kokkinakis, Improved document skew detection based on text line connected-component clustering , 2001 IEEE International Conference on Image Processing, 2001, Vo1.1.1098-1101
[18]J.F.Cullen, K.Ejiri, Weak model-dependent page segmentation and skew correction for processing document images, 1993 IEEE International Conference on Document Analysis and Recognition, 1993.757-760
[19]H.Y.Ma, Z.W.Yu, An enhanced skew angle estimation technique for binary document images, 1999 IEEE International Conference on Document Analysis and Recognition, 1999.165-168
[20]R.Safabakhsh, S.Khadivi, Document skew detection using minimum-area bounding rectangle, 2000 IEEE International Conference on Information Technology, 2000.253-258
[21]王正群.手写体汉字识别研究:[博士学位论文].南京.南京理工大学.2001
[22]丁慧东.脱机手写体汉字识别研究:[硕士学位论文].长春.东北师范大学.2005
[23]冯志敏.基于结构特征的手写体汉字识别研究:[硕士学位论文].昆明.云南师范大学.2006
[24]周光博.一种印刷体汉字特征点提取的新方法: [硕士学位论文].上海.华东师范大学.2006
[25]Romero R D,Touretzky D ,Thibadeaun R H.Optical Chinese character recognition using probabilistic neural networks.Pattern Recognition,1997,30 (8).1279-129.
[26]Cheng Lin Liu,In Jung Kim,Jin H Kim.Model-based stroke extraction and matching for handwritten Chinese character recognition. Pattern Recognition,2001(34).2339-2352
[27]Y Mizukami.A handwritten Chinese character recognition system using Hierarchical displacement extraction based on directional features.Pattern Recognition Letters,1998,(19).595-604
[28]赵明.手写印刷体汉字识别方法综述.计算机研究与发展.1993 ,Vo1.30,No.4,59-64
[29]高彦宇,杨扬.脱机手写体汉字识别研究综述.计算机工程.2004 (7)
[30]路浩如,杨源远.手写体汉字识别问题综述.计算机应用与软件,Vol.l1
[31]丁晓青.汉字识别研究回顾.电子学报.2002 (9)
[32]汪芳,康慕宁等.印刷体汉字识别技术.情报技术.2004 (2)
[33] 柳 回 春 , 马 树 元 , 吴 平 东 等 . 手 写 体 数 字 识 别 技 术 的 研 究 . 计 算 机 工程.2003,29(4):24-25,61
[34]居琰,汪同庆,刘建胜,王贵新,彭健. 基于集成 RBF 神经网络的小类别手写体汉字识别系统.计算机工程与应用.2002.38(23):100-102,158
[35]张世辉,孔令富.汉字识别及现状分析.燕山大学学报. 2003,27(4):367-369
[36]梁涌.印刷体汉字识别系统的研究与实现:[硕士学位论文].西安.西北工业大学. 2006.3
[37]张 鸽,陈书 开.基于SVM的手写 体 阿 拉 伯 数 字 识 别 .军 民 两 用 技 术 与 产品.2005(9):41-43
[38]吴炜,杨晓敏,刘大宇等.一种基于模糊模板匹配的车牌汉字识别方法.微型机与应用.2005,24(11):57-59
[39]崔国伟,舒文豪,李仲荣.多字体多字号印刷汉字识别方法的研究.中文信息学报.1990,4(3):57-62
[40]Nadal C,Legault R,Suen C Y.Complementary algorithms for the recognition of totally unconstrained handwritten numerals[C].In:Proc of the l0th Int Conf Pattern Recognition,1990-06.434-449
[41]Cheng H Yan ,Recogninition of handwritten digits based on contour Information.Pattern Recognition,1998;31(3).235-255
[42]F.H. Cheng and W.H.Hsu,Research on Chinese OCR in Taiwan,Int ,J, Pattern Recognition Artificia1 Intell 5,(1&2).139-164(1991)
[43] C Downton,S Impedovo et a1.Progress in Handwriting Recongnition[M] World Scicentific Publishing Co Pte Ltd, 1997
[44]J Zhou,Q Gan,C Y Suen . A high performance hand-printed numeral recognition system with verification module[C].In:Proc of 4 ICDAR Ulm, Germany, 1997.294-297
[45]王 勇,吴立德.图像恢复和边缘提取的后验均值方法.电子学报.1994,22(2): 70-75
[46]武强,童学锋.季隽.基于人工神经网络的数字字符识别.计算机工程.2003. 8
[47]边肇棋.模式识别.北京:清华大学出版社.1998
[48]Martin T.Hagan, Howard B.Demuth, Mark H.Beale.神经网络设计.北京:机械工业出版社
[49]万红梅,金连文,尹俊勋等.结合距离分类器的神经网络手写体识别.计算机工程与应用,2004. 11
[50]Vapnik V.,The Nature of Statistical leanning Thory.New York: Springer- Verlag,1995
[51]Dong Jian-xiong, Krzyzak A, Suen Ching Y .An improved handwritten Chinese character recognition system using support vector machine .Pattern Recognition Letter.2005,26.1849-1856
[52]石繁槐,童学锋.SVM 在小字符集脱机手写体汉字识别中的应用研究.计算机工程, 2002.6
[53]乔海晔,肖南峰.基于视觉的文字识别系统的设计与实现.交通与计算机,2005(5)
[54]张重阳,娄震,杨静字.基于轮廓和统计特征的手写体数字识别.计算机工程与应用,2004,40(9):83-84,89
[55]娄震,胡钟山,杨静字.基于轮廓分段特征的手写体阿拉伯数字识别.计算机学报, 1999,22(10):1065-1073
[56]郭戈,闫继宏,蒋红梅等.基于结构特征的汉字识别.甘肃工业大学学报. 29(1):81-85