Publications-Theses
Article View/Open
Publication Export
-
Google ScholarTM
NCCU Library
Citation Infomation
Related Publications in TAIR
題名 非監督式新細胞認知機神經網路之研究
Studies on the Unsupervised Neocognitron作者 陳彥勳
Chen, Yen-Shiun貢獻者 蔡瑞煌
Tsaih, Ray-R.
陳彥勳
Chen, Yen-Shiun關鍵詞 神經網路
非監督式學習
新細胞認知機
印刷體中文字辨識
Neural network
Unsupervised learning
Neocognition
Printed chinese character recognition日期 1996 上傳時間 28-Apr-2016 11:55:10 (UTC+8) 摘要 本論文使用非監督式新細胞認知機(Unsupervised neocognitron)神經網路來便是印刷體中文字。
In this study, we are investigating the feasibility of applying the unsupervised neocognitron to the recognition of printed Chinese characters.參考文獻 [1] K. Fukushima, "Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position", BioI Cybern., Vo1.36, pp .193-202,Apr. 1980. [2] Y. LeCun, B. Boser, "Backpropagation Applied to Handwritten Zip Code Recognition", Neural Computation, 1, pp.541-551, 1989. [3] H.Y. Liao, IS. Huang, and S.T. Huang, "Two-Dimentional Neural Networks for Handwritten Chinese Character Recognition", 1992 IEEE IJCNN illS79-S84. [4] A. Rajavelu, M.T. Musavi, and M.V. Shirvaikar, "A Neural Network Approach to Character Recognition", Neural Networks, Vol 2, pp.387-393 1989. [5] K. Fukushima, S. Miyake, T. Ito, "Neocognitron: A Neural Network Model for a Mechanism of Visual Pattern Recognition", IEEE Trans. on System, Man, and Cybernetics, Vol. SMC-13,No.S, Sep/Oct 1983. pp.826-834. [6] K. Fukushima, "Neocognitron: A Hierarchical Neural Network Capable of Visual Pattem Recognition", Neural Networks, Vol.1, pp.1l9-130, 1988. [7] K. Fukushima, N. Wake, "Handwdtten Alphanumeric Character Recognition by the Neocognitron", IEEE Trans. Oll Neural Networks, Vol.2, No.3, May 1991, pp.35S-36S. [8] K. Fukushima, Sei Miyake, "Neocognitron: A New Algorithm For Pattern Recognition Tolerant of Deformation and Shifts In Position", Pattem Recognition, Vol.lS, No.6, pp. 4SS-469,1982. [9] K. Fukushima, N. Wake, "Improved Neocognitroll with Bend-Detecting Cells", Proc. IEEE IJCNN, Vol.4, pp.190-19S, 1992. [10] K. Fukushima, "Analysis of the Process of Visual Pattem Recognition by the Neocognitron",Neural Networks, Vol. 2, pp.413-420, 1989. [11] MuraU M. Menon,Karl G. Heinemann, "Classification of Patterns Using a Self-Organizing Neural Network.", Neural Networks, Vol I, pp.201-21S, 1988. [12] Glenn S. Himes and Rafael M. Inigo, "Automatic Target Recognition Using a Neocognitron",IEEE Trans. on Knowledge and Data Engineering, Vol.4, No.2, April 1992. [13] James A. Freeman, "Neural Networks, Algorit~ Applications, and Program.rrring Techniquesll,Addison-Wesley Publishing Company, July 1992. [14] Hubel, D.H.,Wiesel, T.N.,”Receptive fields, binocular interaction and functional architecture in cat`s visual cortexll, 1. Physiol. 160, pp.l06-1S4, 1962. [IS] Hubel, D.H.,Wiesel, T.N.,"Receptive fields and functional architecture in two nonstriate visual area (18 and 19) of the catll, 1. Neurophysiol. 28,229-289, 1965. [16] Eun Jin Kim, "Handwritten Hangul Recognition Using a Modified Neocognitron", Neural Networks, Vol 4, pp.743-7S0, 1991. [17] S. Yamaguchi, H. Itakura, "A Car Detection System Using the Neocognitron", Proc. IEEE IJCNN, Vol.2, pp.1208-1213, 1991. [18] S.D. Wang, C.C. Pan, "A Neural Network Approach for Chinese Character Recognition", Proc.IEEE IJCNN, Vol.1, pp.416-419, 1990. [19] F.G. Shieh, “Studies of the Recognition of the Printed Chinese Character Using the NeocognitfOn model with the Changjei Codes", Master thesis of Computer and Information Engineering, Tatung Institute of Engineering, July 1993. 描述 碩士
國立政治大學
資訊管理學系
83356004資料來源 http://thesis.lib.nccu.edu.tw/record/#B2002002866 資料類型 thesis dc.contributor.advisor 蔡瑞煌 zh_TW dc.contributor.advisor Tsaih, Ray-R. en_US dc.contributor.author (Authors) 陳彥勳 zh_TW dc.contributor.author (Authors) Chen, Yen-Shiun en_US dc.creator (作者) 陳彥勳 zh_TW dc.creator (作者) Chen, Yen-Shiun en_US dc.date (日期) 1996 en_US dc.date.accessioned 28-Apr-2016 11:55:10 (UTC+8) - dc.date.available 28-Apr-2016 11:55:10 (UTC+8) - dc.date.issued (上傳時間) 28-Apr-2016 11:55:10 (UTC+8) - dc.identifier (Other Identifiers) B2002002866 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/87347 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 資訊管理學系 zh_TW dc.description (描述) 83356004 zh_TW dc.description.abstract (摘要) 本論文使用非監督式新細胞認知機(Unsupervised neocognitron)神經網路來便是印刷體中文字。 zh_TW dc.description.abstract (摘要) In this study, we are investigating the feasibility of applying the unsupervised neocognitron to the recognition of printed Chinese characters. en_US dc.description.tableofcontents ABSTRACT(in Chinese)..........i ABSTRACT(in English)..........ii CONTENTS..........iv LIST OF TABLES..........vi LIST OF FIGURES..........viii Chapter 1 Introduction..........1 1.1Motivation..........1 1.2Research topics and goal..........3 1.3The empirical way of analysis..........3 1.4Thesis organization..........4 Chapter 2 Neocognitron..........5 2.1Structure of the network..........5 2.2Behavior of cells..........9 2.2.1Feature extraction by an S-cell..........10 2.2.2The role of C-cell..........12 2.3Unsupervised learning..........13 2.4Review of earlier studies..........15 2.4.1Earlier work on the unsupervised neocognitron..........15 2.4.2Summary of Literature review..........19 Chapter 3 Experiments of Part I..........20 3.1The propositions for the unsupervised neocognitron..........21 3.2The ways of training and testing in experiments of Part I..........21 3.2.1Training process..........22 3.2.2Testing process..........22 3.3Experiment 1..........23 3.4Experiment 2..........24 3.5Experiment 3..........28 3.6The reasons for the divergence of the neocognitron..........31 3.7Experiment 4..........37 Chapter 4 Experiments of Part II..........44 4.1Experiment 1..........44 4.1.1The successful learning rate..........44 4.1.2The recognition rate..........47 4.2Experiment 2..........49 4.2.1Simulation results for SYSTEM A..........49 4.2.2Simulation results for SYSTEM B..........52 Chapter 5 Summary and future work..........54 5.1Summary..........54 5.2Future work..........55 References..........57 Appendix A..........59 Appendix B..........62 Appendix C..........64 zh_TW dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#B2002002866 en_US dc.subject (關鍵詞) 神經網路 zh_TW dc.subject (關鍵詞) 非監督式學習 zh_TW dc.subject (關鍵詞) 新細胞認知機 zh_TW dc.subject (關鍵詞) 印刷體中文字辨識 zh_TW dc.subject (關鍵詞) Neural network en_US dc.subject (關鍵詞) Unsupervised learning en_US dc.subject (關鍵詞) Neocognition en_US dc.subject (關鍵詞) Printed chinese character recognition en_US dc.title (題名) 非監督式新細胞認知機神經網路之研究 zh_TW dc.title (題名) Studies on the Unsupervised Neocognitron en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) [1] K. Fukushima, "Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position", BioI Cybern., Vo1.36, pp .193-202,Apr. 1980. [2] Y. LeCun, B. Boser, "Backpropagation Applied to Handwritten Zip Code Recognition", Neural Computation, 1, pp.541-551, 1989. [3] H.Y. Liao, IS. Huang, and S.T. Huang, "Two-Dimentional Neural Networks for Handwritten Chinese Character Recognition", 1992 IEEE IJCNN illS79-S84. [4] A. Rajavelu, M.T. Musavi, and M.V. Shirvaikar, "A Neural Network Approach to Character Recognition", Neural Networks, Vol 2, pp.387-393 1989. [5] K. Fukushima, S. Miyake, T. Ito, "Neocognitron: A Neural Network Model for a Mechanism of Visual Pattern Recognition", IEEE Trans. on System, Man, and Cybernetics, Vol. SMC-13,No.S, Sep/Oct 1983. pp.826-834. [6] K. Fukushima, "Neocognitron: A Hierarchical Neural Network Capable of Visual Pattem Recognition", Neural Networks, Vol.1, pp.1l9-130, 1988. [7] K. Fukushima, N. Wake, "Handwdtten Alphanumeric Character Recognition by the Neocognitron", IEEE Trans. Oll Neural Networks, Vol.2, No.3, May 1991, pp.35S-36S. [8] K. Fukushima, Sei Miyake, "Neocognitron: A New Algorithm For Pattern Recognition Tolerant of Deformation and Shifts In Position", Pattem Recognition, Vol.lS, No.6, pp. 4SS-469,1982. [9] K. Fukushima, N. Wake, "Improved Neocognitroll with Bend-Detecting Cells", Proc. IEEE IJCNN, Vol.4, pp.190-19S, 1992. [10] K. Fukushima, "Analysis of the Process of Visual Pattem Recognition by the Neocognitron",Neural Networks, Vol. 2, pp.413-420, 1989. [11] MuraU M. Menon,Karl G. Heinemann, "Classification of Patterns Using a Self-Organizing Neural Network.", Neural Networks, Vol I, pp.201-21S, 1988. [12] Glenn S. Himes and Rafael M. Inigo, "Automatic Target Recognition Using a Neocognitron",IEEE Trans. on Knowledge and Data Engineering, Vol.4, No.2, April 1992. [13] James A. Freeman, "Neural Networks, Algorit~ Applications, and Program.rrring Techniquesll,Addison-Wesley Publishing Company, July 1992. [14] Hubel, D.H.,Wiesel, T.N.,”Receptive fields, binocular interaction and functional architecture in cat`s visual cortexll, 1. Physiol. 160, pp.l06-1S4, 1962. [IS] Hubel, D.H.,Wiesel, T.N.,"Receptive fields and functional architecture in two nonstriate visual area (18 and 19) of the catll, 1. Neurophysiol. 28,229-289, 1965. [16] Eun Jin Kim, "Handwritten Hangul Recognition Using a Modified Neocognitron", Neural Networks, Vol 4, pp.743-7S0, 1991. [17] S. Yamaguchi, H. Itakura, "A Car Detection System Using the Neocognitron", Proc. IEEE IJCNN, Vol.2, pp.1208-1213, 1991. [18] S.D. Wang, C.C. Pan, "A Neural Network Approach for Chinese Character Recognition", Proc.IEEE IJCNN, Vol.1, pp.416-419, 1990. [19] F.G. Shieh, “Studies of the Recognition of the Printed Chinese Character Using the NeocognitfOn model with the Changjei Codes", Master thesis of Computer and Information Engineering, Tatung Institute of Engineering, July 1993. zh_TW