基于多分类器集成的模式识别研究
|
摘要
在人工智能领域，模式识别是一个非常重要的方面。本文就如何提高识别性能对模式识别系统进行了研究。
本文首先对基于BP学习算法的神经网络模式识别方法进行分析和研究，指出了其存在的缺点，利用模糊逻辑的知识表达能力，在神经网络中引入模糊逻辑，构成模糊神经网络来改善网络性能，针对模糊神经网络的参数难以选择问题，提出了基于遗传算法的模糊神经网络学习模型。但是单一分类器由于其采用的特征类型单一性以及自身的局限性，改进的网络模型虽然在一定程度上提高了识别性能，但是这种提高有限。
研究发现，不同分类器在识别性能上有互补作用，因此如何把各种分类器结合在一起，从而能够集成各个分类器的优点，而抑制它们的缺点，是提高识别性能的关键，对该问题的研究，虽然已经取得了一些成果，但是研究主要集中于抽象级信息的集成，而忽略了表征分类器性能的完整信息——度量级信息，因此不能从根本上解决分类器性能的互补问题。
针对上述问题，本文对多分类器集成进行了深入研究。首先总结了用于模式识别的多分类器集成机理以及存在的问题，并对各种基于抽象级信息的建模方法进行分析和比较，指出了基于抽象级信息的分类器集成存在的缺点，提出了基于度量级信息的分类器集成，并采用模糊积分进行集成建模，对于模型中的模糊积分密度难以确定问题，采用了一种动态评价方法——贝叶斯方法，仿真结果表明了该建模方法的有效性。
As an important aspect in the domain of artificial intelligence, pattern recognition can extend the application range of computer and improve the ability of computers to perceive outside information. In this paper, research about how to improve recognition performance is presented.
In the thesis, neural network pattern recognition based on BP algorithm is analysised and researched , and its disadvantage is pointed out. With the knowledge representation ability of fuzzy logic, we use fuzzy logic in neural network, and build up fuzzy neural network to improve network performance. Furthermore, We develop a kind of fuzzy neural network learning model based on Generic algorithm to solve the problem that how to decide the parameters of network and better performance can be achieved, while single classifier use single feature and has its limitation, improved network can't be get expected result.
It's suggested that different classifier offered complementary information, which motivated the interest in combining classifiers to harness their strength. Some achievements are obtained, however most research focused on combination based abstract information and ignored the combination based on measured information, so complementary problem of classifiers hasn't been solved completely.
for the above problem, a deep research about the combination of multiple classifier is presented, at first, we conclude the principle of combination multiple classifier and existed problem, and compare the different methods based on abstract information, also their drawback is pointed out. Then combination of measure information is described with fuzzy integral, and a dynamic evaluated method, bayes method, is to decide fuzzy integral density of model. Simulation shows efficiency of the method
本文首先对基于BP学习算法的神经网络模式识别方法进行分析和研究，指出了其存在的缺点，利用模糊逻辑的知识表达能力，在神经网络中引入模糊逻辑，构成模糊神经网络来改善网络性能，针对模糊神经网络的参数难以选择问题，提出了基于遗传算法的模糊神经网络学习模型。但是单一分类器由于其采用的特征类型单一性以及自身的局限性，改进的网络模型虽然在一定程度上提高了识别性能，但是这种提高有限。
研究发现，不同分类器在识别性能上有互补作用，因此如何把各种分类器结合在一起，从而能够集成各个分类器的优点，而抑制它们的缺点，是提高识别性能的关键，对该问题的研究，虽然已经取得了一些成果，但是研究主要集中于抽象级信息的集成，而忽略了表征分类器性能的完整信息——度量级信息，因此不能从根本上解决分类器性能的互补问题。
针对上述问题，本文对多分类器集成进行了深入研究。首先总结了用于模式识别的多分类器集成机理以及存在的问题，并对各种基于抽象级信息的建模方法进行分析和比较，指出了基于抽象级信息的分类器集成存在的缺点，提出了基于度量级信息的分类器集成，并采用模糊积分进行集成建模，对于模型中的模糊积分密度难以确定问题，采用了一种动态评价方法——贝叶斯方法，仿真结果表明了该建模方法的有效性。
As an important aspect in the domain of artificial intelligence, pattern recognition can extend the application range of computer and improve the ability of computers to perceive outside information. In this paper, research about how to improve recognition performance is presented.
In the thesis, neural network pattern recognition based on BP algorithm is analysised and researched , and its disadvantage is pointed out. With the knowledge representation ability of fuzzy logic, we use fuzzy logic in neural network, and build up fuzzy neural network to improve network performance. Furthermore, We develop a kind of fuzzy neural network learning model based on Generic algorithm to solve the problem that how to decide the parameters of network and better performance can be achieved, while single classifier use single feature and has its limitation, improved network can't be get expected result.
It's suggested that different classifier offered complementary information, which motivated the interest in combining classifiers to harness their strength. Some achievements are obtained, however most research focused on combination based abstract information and ignored the combination based on measured information, so complementary problem of classifiers hasn't been solved completely.
for the above problem, a deep research about the combination of multiple classifier is presented, at first, we conclude the principle of combination multiple classifier and existed problem, and compare the different methods based on abstract information, also their drawback is pointed out. Then combination of measure information is described with fuzzy integral, and a dynamic evaluated method, bayes method, is to decide fuzzy integral density of model. Simulation shows efficiency of the method
引文
Ajantha S.Atukorale, P.N.Suganthan, Multiple hong network fusion by fuzzy integral All. K. M, Pazzani. M. .J, On the link between error correlation and error reduction in decision tree ensembles, Technical report 95-38, ICS-UCI, 1995
Bajaj R, Chaudhury S, Signature verification using multiple neural classifiers, Pattern recognition, 1997, 30(1) : 7-7
Barnett. J. A, Computional methods for a mathematical theory of evidence, Proc 7th Int Joint Conf Arlifical Intell, Vancouver DC, 1985:868-875
Brown. R. M, Fay. T.H ct al, Handprinted symbol recognition system, Pattern Recognition, 1988 21(2) : 91-118
Chan. L. W, Fallside. F, An adaptive training algorithm for back-progagation networks, Computer Speech and Language, 1987 2
Ching Y. Suen, Computer Recognition of Unconstrained Handwritten Numerals, Proceedings of the IEEE, 1992 80(7) : 1162-1179
Cordelia. L. P, Foggia. P et al, Reliability parameters to improve combination strategies in multi-expert systems, Pattern Analysis & Application ,1999 2: 205-214
Cun. Y. Le et al, Constrained neural network for unconstrained handwritten digit recognition, Proc Int Workshop Frontiers in Handwritting recognition(Concordia University, Montreal), Apr. 1990, pp: 145-154
David M.J. Tax, Martijn Van Breukelen et al, Combining multiple classifiers by averaging or by mulitplying, Pattern recognition 2000(33) : 1475-1485
Dilecce. V, Dimauro. G et al, Classifier combination: the role of a-priori konwledge Holland JH, Adaptation in natural and artificial systems, Univ of Michigan Press, Ann Arbor Mich, 1975
Huang. Y. S, Suen. C. Y, A method of Combining multiple experts for the recognition of unconrstrained handwritten numerals, IEEE Trans on PA&MI, 1995 17(1) :90-93
Hull. J. J, Commike. A, Ho. T. K, Multiple algorithms for handwritten character recogniton, Proc Int workshop frontiers in handwriting recognition, Montreal, PQ, Canada, 1990: 117-129
Jinhai Cai, Zhi-Qiang Liu, Integration of structural and statistical information for unconstrained handwritten numeral recognition, IEEE Trans on Pattern analysis and machine intelligence, 1999 21(3) :263-270
Josef Kittler, Mohamad et al, On combining classifiers IEEE Trans on PA&MI 1998 20(3) :226-239
Jun Cao et al, Recognition of Handwritten Numerals with Multiple Featture and Multistage Classifer, Pattern recogniton, 199528(2) : 153-160Kosko B, Fuzzy associative memories, In: Kondel A(ed). Fuzzy Experts System Reading. MA: Addison Weley, 1987
Kittler J, Improving Recognition Rates by Classifier Combination: A Theortical Framework, In: Downtown AC, Impedovos(eds), Progress in Handwriting Recognition. World Scientific, 1997, pp: 231-248
Klir. G. J, Wang. Z.Y. et al, Constructing fuzzy measures in expert system, Fuzzy sets and system, 1997 9(2) : 251-264
Kosko. B, Neural Networks and Fuzzy Systems: A Dynamical Systems Approach to Machine Intelligence, Englewood Cliffs, NJ: Prentice-Hall, 1992
Kwan C. C, Pang L, Suen C. Y, A Comparative Study of some Recognition Algorithms in Character Recognition, Proc. Int. Conf. On Cybernetice and Society (Denver), 1979pp: 530-535
Kwan H K, Cai Yaling, A Fuzzy Neural Network and Its Application to Pattern Recognition, IEEE Trans FS, 1994 2 (3) : 185-193
Kwok-Ping Chan, Learning templates from fuzzy examples in structural recognition, IEEE Trans on SYC-Part B, 26(1) : 118-123
Mitchell. B. T, Gillies. A. M, A model-based computer vision system for handwritten ZIP codes. Machine Vision and Application, 1989 2: 231-243
Michael D. Garris, Charles L. Wilson, Neural network-based systems for handprint OCR application, IEEE Tram on images processing, 1998 7(8) : 1097-1111
Nadal. C, Legault. R, Suen. C. Y, Complementary algorithms for recognition of totally unconstrained handwritten numerals, Proc 10th Int. Conf. Pattern Recognition 1990 A pp: 434-449
Nabil. Jean. Naccache, SPTA: A proposed algorithm for thinning binary patterns, IEEE Tram on SMC, 1984 14(3) : 409-417
Oivid. Due, Trier et al, Feature extraction methods for character recognition-a survey , Pattern Recognition, 1996 29(4) : 641-661
Pfeffer. W. F, Integrals and Measures New York: Marcel Dekker 1977 Pudil P, Multistage Pattern Recognition with Reject Option, Proc 11th IAPR ICPR 1992: 92-95
Rahman A. F. R, Fairhurst M.C. , Serial combination of multiple experts: a unifed valuation, Pattern Analysis & Application , 1999 2: 292-311
Rarndas V, Sridhar V et al, An effective technique for feature extraction, Pattern Recognition Letters, 1994 15: 885-891
Ranachari Anand et al, Efficient classification for mulitcalss problem using modular neural networks, IEEE Trans on neural networks, 1995 6(1) : 117-124
Richard D. Romero, David S. Tourtzky et al, .Pattern recognition, 1997 30(8) : 1279-1292
Schaffer .TD, Whitley D, Proceedings of the Workshop on Combinations of Genetic Algorithms and Neural Networks, Los Alamitos,CA:The IEEE Computer Society Press. 1992
Simpson P.K, Fuzzy min-max neural network-part 1:classification, IEEE Trans NN 1992 3(5): 776-786
Stringa. L, Effifient classification of totally unconstrained handwritten numerals with a trainable multilayer network, Pattern Recognition Left, 1989 10: 273-280
Sugeno. M, Fuzzy measures and fuzzy integrals: A survey Fuzzy Automata and decision, Processes Amsterdam: North Holland I977: 89-102
Sung Bae Cho, Jin H.Kim, Combining multiple networks by fuzzy integral for Robust Classification, IEEE Trans on SMC 1995 25(2): 380-384
Sung Bae Cho, Neural-network classifiers for recognizing totally unconstrained handwritten numerals, IEEE Trans on neural nerworks 1997 8(1): 43-53
Tahani H, Keller JM, Information Fusion in Computer Vision Using the Fuzzy Integral, IEEE Trans on SMC, 1990, 20(3): 733～741
Tin Kam Ho, Decision Combination in Multiple Classifier Systems, IEEE Trans on PA & MI, 94 16(1) PP: 66-75
Unger S H, Pattern Detection and Recognition, Proceedings of the IRE, 1959, 1737-1752
Verma B., P. Gader, Fusion of multiple handwritten word recognition techniques, IEEE 2000
Wang Zhenyuan, Klir, G. J., Fuzzy measure theory, Plenum Press, New York, 1992
Wierzechon. S. T, On fuzzy measure and fuzzy integral, Fuzzy information and decision processes
Xu. Lei et al, Methods of Combining Multiple Classifiers and Their Application to Handwriting Recognition. IEEE Trans on System, Man And Cybernetics 1992(22): 418-435
Yager R R, Element Selection from a Fuzzy Subset Using the Fuzzy Integral, IEEE Trans on SMC, 1993, 23(2): 467-477
Yamakawa T, Tomoda S, A Fuzzy Neuron and Its Application to Pattern Recognition. Proc Third Int Fuzzy System Associat Congress. 1989: 30-38
Zadeh. L, Fuzzy sets, Inform. and control. 1965 8:338-353
边肇祺 张学工等，模式识别(第二版)，北京：清华大学出版社 2000
戴汝为 郝红卫，综合集成的构思在模式识别中的应用，自动化学报，1997 23(3)：302-307
段新生，证据理论与决策、人工智能，北京：中国人民大学出版社 1993
丁震，胡钟山，杨静宇等，一种基于模糊聚类的图象分割方法，计算机研究与发展，1997 34(7)：536-541
费文东，孟相如 细胞神经网络与字符特征提取技术研究空军工程大学学报(自然科学版)2000 1(3)：51-54
郭桂容，庄钊文 信息处理中模糊技术 长沙：国防科技大学出版社 1993
傅京孙主编，模式识别应用，程民德等译，北京：北京大学出版社，1990
胡钟山 娄震等，基于多分类器组合的手写数字识别，计算机学报，1999 22(4)：369-374
胡敏瑞 等，人工神经网络的智能神经元模型，电子学报 1996 24(4)：86-90
黄德双 著，神经网络模式识别系统理论，北京：电子工业出版社，1996
金连文，彭秀兰，尹俊勋，一种手写体汉字特征提取新方法——小波变换及弹性网格，
李晶皎，孙杰，姚天顺，语音识别中基于SFCM模糊聚类的矢量量化方法，计算机研究与发展，1999 36(3)：264-267
李敏强，徐博艺，寇纪淞，遗传算法与神经网络的结合，系统工程理论与实践，1999 2：65-69
刘普寅，张汉江，吴孟达等，模糊神经网络理论研究综述，模糊系统和数学，199812(1)：77-87
令狐选霞 等，一种新的改进遗传算法——混合式遗传算法，系统工程与电子技术，2001 23(7)：95-97
吕岳 施鹏飞等，改进的贝叶斯多分类器组合规则，数据采集与处理，2000 15(2)：204-207
马少平 夏莹 等，汉字的层次轮廓特征及其应用，清华大学学报(自然科学版)，1995 35(5)：79-83
马少平，夏莹，朱小燕，基于模糊方向线素特征的手写体汉字识别清华大学学报(自然科学版)，1997，37(3)：42-45
瞿东晖，张立明 多层前馈网络在模式识别中的理论和应用 1995 23(7)：64-68
任平，遗传算法(综述)，工程数学学报，1999，16(1)
孙立民 狄红卫等，基于子块及其相关模糊特征的手写体汉字识别方法，通信学报，1999 20(12)：81-85
宋红萍，刘宏超等基于小波网络和多模块网络的数字识别，中文信息学报
孙增圻 等 智能控制理论与技术，北京：清华大学出版社 1997
孙怀江，胡钟山，杨静宇等，基于证据理论的多分类器融合方法研究，计算机学报，2001 24(3)：231-235
朱小燕等，手写体字符识别研究，模式识别与人工智能 2000，13(2)：174-180技术的应用，1998 3(7)：549-552
张德喜，基于模糊线素特征与神经网络相结合的手写体汉字识别，许昌师专学报，2000，19(2)：59-63
王士同 著，神经模糊系统及其应用，北京：北京航空航天大学出版社，1998
王永庆 人工智能原理与方法 西安：西安交通大学出版社 1998
王煦法，遗传算法及其应用，小型微型计算机系统，1995 16(2)：59-64
王震源，李法朝，Fuzzy积分在评判过程中的应用，模糊数学，19851：109-114
王正群，叶晖 等，基于模糊方向特征的手写体汉字识别，模式识别与人工智能，2001 14(3)：317-320
吴从炘，马明，模糊分析学基础 北京：国防工业出版社 1991
席裕庚 等，遗传算法综述，控制理论与应用，1996 13(6)：697-708
恽为民 基于遗传的机器人运动规划，上海交通大学博士论文，1995
张文修，模糊数学引论 西安：西安交通大学出版社 1991
张炘中，汉字识别技术，北京：清华大学出版社 1992
Bajaj R, Chaudhury S, Signature verification using multiple neural classifiers, Pattern recognition, 1997, 30(1) : 7-7
Barnett. J. A, Computional methods for a mathematical theory of evidence, Proc 7th Int Joint Conf Arlifical Intell, Vancouver DC, 1985:868-875
Brown. R. M, Fay. T.H ct al, Handprinted symbol recognition system, Pattern Recognition, 1988 21(2) : 91-118
Chan. L. W, Fallside. F, An adaptive training algorithm for back-progagation networks, Computer Speech and Language, 1987 2
Ching Y. Suen, Computer Recognition of Unconstrained Handwritten Numerals, Proceedings of the IEEE, 1992 80(7) : 1162-1179
Cordelia. L. P, Foggia. P et al, Reliability parameters to improve combination strategies in multi-expert systems, Pattern Analysis & Application ,1999 2: 205-214
Cun. Y. Le et al, Constrained neural network for unconstrained handwritten digit recognition, Proc Int Workshop Frontiers in Handwritting recognition(Concordia University, Montreal), Apr. 1990, pp: 145-154
David M.J. Tax, Martijn Van Breukelen et al, Combining multiple classifiers by averaging or by mulitplying, Pattern recognition 2000(33) : 1475-1485
Dilecce. V, Dimauro. G et al, Classifier combination: the role of a-priori konwledge Holland JH, Adaptation in natural and artificial systems, Univ of Michigan Press, Ann Arbor Mich, 1975
Huang. Y. S, Suen. C. Y, A method of Combining multiple experts for the recognition of unconrstrained handwritten numerals, IEEE Trans on PA&MI, 1995 17(1) :90-93
Hull. J. J, Commike. A, Ho. T. K, Multiple algorithms for handwritten character recogniton, Proc Int workshop frontiers in handwriting recognition, Montreal, PQ, Canada, 1990: 117-129
Jinhai Cai, Zhi-Qiang Liu, Integration of structural and statistical information for unconstrained handwritten numeral recognition, IEEE Trans on Pattern analysis and machine intelligence, 1999 21(3) :263-270
Josef Kittler, Mohamad et al, On combining classifiers IEEE Trans on PA&MI 1998 20(3) :226-239
Jun Cao et al, Recognition of Handwritten Numerals with Multiple Featture and Multistage Classifer, Pattern recogniton, 199528(2) : 153-160Kosko B, Fuzzy associative memories, In: Kondel A(ed). Fuzzy Experts System Reading. MA: Addison Weley, 1987
Kittler J, Improving Recognition Rates by Classifier Combination: A Theortical Framework, In: Downtown AC, Impedovos(eds), Progress in Handwriting Recognition. World Scientific, 1997, pp: 231-248
Klir. G. J, Wang. Z.Y. et al, Constructing fuzzy measures in expert system, Fuzzy sets and system, 1997 9(2) : 251-264
Kosko. B, Neural Networks and Fuzzy Systems: A Dynamical Systems Approach to Machine Intelligence, Englewood Cliffs, NJ: Prentice-Hall, 1992
Kwan C. C, Pang L, Suen C. Y, A Comparative Study of some Recognition Algorithms in Character Recognition, Proc. Int. Conf. On Cybernetice and Society (Denver), 1979pp: 530-535
Kwan H K, Cai Yaling, A Fuzzy Neural Network and Its Application to Pattern Recognition, IEEE Trans FS, 1994 2 (3) : 185-193
Kwok-Ping Chan, Learning templates from fuzzy examples in structural recognition, IEEE Trans on SYC-Part B, 26(1) : 118-123
Mitchell. B. T, Gillies. A. M, A model-based computer vision system for handwritten ZIP codes. Machine Vision and Application, 1989 2: 231-243
Michael D. Garris, Charles L. Wilson, Neural network-based systems for handprint OCR application, IEEE Tram on images processing, 1998 7(8) : 1097-1111
Nadal. C, Legault. R, Suen. C. Y, Complementary algorithms for recognition of totally unconstrained handwritten numerals, Proc 10th Int. Conf. Pattern Recognition 1990 A pp: 434-449
Nabil. Jean. Naccache, SPTA: A proposed algorithm for thinning binary patterns, IEEE Tram on SMC, 1984 14(3) : 409-417
Oivid. Due, Trier et al, Feature extraction methods for character recognition-a survey , Pattern Recognition, 1996 29(4) : 641-661
Pfeffer. W. F, Integrals and Measures New York: Marcel Dekker 1977 Pudil P, Multistage Pattern Recognition with Reject Option, Proc 11th IAPR ICPR 1992: 92-95
Rahman A. F. R, Fairhurst M.C. , Serial combination of multiple experts: a unifed valuation, Pattern Analysis & Application , 1999 2: 292-311
Rarndas V, Sridhar V et al, An effective technique for feature extraction, Pattern Recognition Letters, 1994 15: 885-891
Ranachari Anand et al, Efficient classification for mulitcalss problem using modular neural networks, IEEE Trans on neural networks, 1995 6(1) : 117-124
Richard D. Romero, David S. Tourtzky et al, .Pattern recognition, 1997 30(8) : 1279-1292
Schaffer .TD, Whitley D, Proceedings of the Workshop on Combinations of Genetic Algorithms and Neural Networks, Los Alamitos,CA:The IEEE Computer Society Press. 1992
Simpson P.K, Fuzzy min-max neural network-part 1:classification, IEEE Trans NN 1992 3(5): 776-786
Stringa. L, Effifient classification of totally unconstrained handwritten numerals with a trainable multilayer network, Pattern Recognition Left, 1989 10: 273-280
Sugeno. M, Fuzzy measures and fuzzy integrals: A survey Fuzzy Automata and decision, Processes Amsterdam: North Holland I977: 89-102
Sung Bae Cho, Jin H.Kim, Combining multiple networks by fuzzy integral for Robust Classification, IEEE Trans on SMC 1995 25(2): 380-384
Sung Bae Cho, Neural-network classifiers for recognizing totally unconstrained handwritten numerals, IEEE Trans on neural nerworks 1997 8(1): 43-53
Tahani H, Keller JM, Information Fusion in Computer Vision Using the Fuzzy Integral, IEEE Trans on SMC, 1990, 20(3): 733～741
Tin Kam Ho, Decision Combination in Multiple Classifier Systems, IEEE Trans on PA & MI, 94 16(1) PP: 66-75
Unger S H, Pattern Detection and Recognition, Proceedings of the IRE, 1959, 1737-1752
Verma B., P. Gader, Fusion of multiple handwritten word recognition techniques, IEEE 2000
Wang Zhenyuan, Klir, G. J., Fuzzy measure theory, Plenum Press, New York, 1992
Wierzechon. S. T, On fuzzy measure and fuzzy integral, Fuzzy information and decision processes
Xu. Lei et al, Methods of Combining Multiple Classifiers and Their Application to Handwriting Recognition. IEEE Trans on System, Man And Cybernetics 1992(22): 418-435
Yager R R, Element Selection from a Fuzzy Subset Using the Fuzzy Integral, IEEE Trans on SMC, 1993, 23(2): 467-477
Yamakawa T, Tomoda S, A Fuzzy Neuron and Its Application to Pattern Recognition. Proc Third Int Fuzzy System Associat Congress. 1989: 30-38
Zadeh. L, Fuzzy sets, Inform. and control. 1965 8:338-353
边肇祺 张学工等，模式识别(第二版)，北京：清华大学出版社 2000
戴汝为 郝红卫，综合集成的构思在模式识别中的应用，自动化学报，1997 23(3)：302-307
段新生，证据理论与决策、人工智能，北京：中国人民大学出版社 1993
丁震，胡钟山，杨静宇等，一种基于模糊聚类的图象分割方法，计算机研究与发展，1997 34(7)：536-541
费文东，孟相如 细胞神经网络与字符特征提取技术研究空军工程大学学报(自然科学版)2000 1(3)：51-54
郭桂容，庄钊文 信息处理中模糊技术 长沙：国防科技大学出版社 1993
傅京孙主编，模式识别应用，程民德等译，北京：北京大学出版社，1990
胡钟山 娄震等，基于多分类器组合的手写数字识别，计算机学报，1999 22(4)：369-374
胡敏瑞 等，人工神经网络的智能神经元模型，电子学报 1996 24(4)：86-90
黄德双 著，神经网络模式识别系统理论，北京：电子工业出版社，1996
金连文，彭秀兰，尹俊勋，一种手写体汉字特征提取新方法——小波变换及弹性网格，
李晶皎，孙杰，姚天顺，语音识别中基于SFCM模糊聚类的矢量量化方法，计算机研究与发展，1999 36(3)：264-267
李敏强，徐博艺，寇纪淞，遗传算法与神经网络的结合，系统工程理论与实践，1999 2：65-69
刘普寅，张汉江，吴孟达等，模糊神经网络理论研究综述，模糊系统和数学，199812(1)：77-87
令狐选霞 等，一种新的改进遗传算法——混合式遗传算法，系统工程与电子技术，2001 23(7)：95-97
吕岳 施鹏飞等，改进的贝叶斯多分类器组合规则，数据采集与处理，2000 15(2)：204-207
马少平 夏莹 等，汉字的层次轮廓特征及其应用，清华大学学报(自然科学版)，1995 35(5)：79-83
马少平，夏莹，朱小燕，基于模糊方向线素特征的手写体汉字识别清华大学学报(自然科学版)，1997，37(3)：42-45
瞿东晖，张立明 多层前馈网络在模式识别中的理论和应用 1995 23(7)：64-68
任平，遗传算法(综述)，工程数学学报，1999，16(1)
孙立民 狄红卫等，基于子块及其相关模糊特征的手写体汉字识别方法，通信学报，1999 20(12)：81-85
宋红萍，刘宏超等基于小波网络和多模块网络的数字识别，中文信息学报
孙增圻 等 智能控制理论与技术，北京：清华大学出版社 1997
孙怀江，胡钟山，杨静宇等，基于证据理论的多分类器融合方法研究，计算机学报，2001 24(3)：231-235
朱小燕等，手写体字符识别研究，模式识别与人工智能 2000，13(2)：174-180技术的应用，1998 3(7)：549-552
张德喜，基于模糊线素特征与神经网络相结合的手写体汉字识别，许昌师专学报，2000，19(2)：59-63
王士同 著，神经模糊系统及其应用，北京：北京航空航天大学出版社，1998
王永庆 人工智能原理与方法 西安：西安交通大学出版社 1998
王煦法，遗传算法及其应用，小型微型计算机系统，1995 16(2)：59-64
王震源，李法朝，Fuzzy积分在评判过程中的应用，模糊数学，19851：109-114
王正群，叶晖 等，基于模糊方向特征的手写体汉字识别，模式识别与人工智能，2001 14(3)：317-320
吴从炘，马明，模糊分析学基础 北京：国防工业出版社 1991
席裕庚 等，遗传算法综述，控制理论与应用，1996 13(6)：697-708
恽为民 基于遗传的机器人运动规划，上海交通大学博士论文，1995
张文修，模糊数学引论 西安：西安交通大学出版社 1991
张炘中，汉字识别技术，北京：清华大学出版社 1992