跳到主要內容

臺灣博碩士論文加值系統

(44.222.189.51) 您好!臺灣時間:2024/05/18 18:55
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:林義貴
研究生(外文):Yi-Guei Lin
論文名稱:以支撐向量機為基礎之影像多值化處理及手寫中文辨識
論文名稱(外文):Multilevel Image Thresholding and Handwritten Chinese Character Recognition by Support Vector Machines
指導教授:董呈煌董呈煌引用關係
指導教授(外文):Cheng-Huang Tung
學位類別:碩士
校院名稱:國立屏東商業技術學院
系所名稱:資訊工程系(所)
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2011
畢業學年度:99
語文別:中文
論文頁數:118
中文關鍵詞:影像二值化影像多階層化支撐向量機手寫中文辨識
外文關鍵詞:binarization thresholdinmultilevel image thresholdinSVMhandwritten Chinese character recognition
相關次數:
  • 被引用被引用:1
  • 點閱點閱:561
  • 評分評分:
  • 下載下載:37
  • 收藏至我的研究室書目清單書目收藏:1
近年來支撐向量機(Support Vector Machines)在分類應用的研究領域倍受重視,例如手寫字元辨識(Handwritten Character Recognition)、人臉辨識(Face Recognition)、文字分類(Text Categorization)…等。在這些既有研究的實驗數據中,支撐向量機均有相當好的成果,其效能因此受到肯定;而SVM所使用的解最佳化問題的數學原理,涉及的是非線性規劃(Nonlinear Programming)中的二次規劃(Quadratic Programming)問題,也有許多機器學習領域的研究進行探討,以求能進一步提升支撐向量機的解題效能。

本論文將先提出以SVM為基礎的影像二值化(Bi-Level Image Thresholding)法,此方法可以有效利用被選定的灰階影像的像素點之特徵(包含平面座標值和其它特徵,如灰階值和Gradient值),成功的對SVM訓練後,再以該SVM對該灰階影像的所有像素點進行二值化分類,根據實驗結果顯示,本論文所提出的二值化法,特別對明暗分布不均的灰階影像有非常好的二值化效果。本論文接續以所提出的二值化法為基礎,進一步提出以SVM為基礎的影像多階層化(Multilevel Image Thresholding)法,其中使用遞迴的方式,將一灰階影像有效的實現所需求的影像多階層化動作,實驗結果顯示,所提出的影像多階層化法具有可將具同一灰階值的諸多影像點變動分類的能力,使處理所得的多階層影像具有較佳的效果。除此之外,本研究也利用SVM來辨識中文字形,初步先擷取中文字形的特徵,運用簡單的Mean分類後,再使用SVM來做分類,實驗結果顯示,整體的中文辨識率最高可以達到98.31%。
Recently, support vector machines (SVM) has been adopted in various categorization applications, such as handwritten character recognition, face recognition, and text categorization. The experimental results of these researches confirmed the effectiveness of SVM. The mathematical theory used in SVM is related to the subfield of nonlinear programming, quadratic programming, which has been developed by many researches in the machine learning area to forther improve the performance of SVM.

The research first proposed a SVM-based bi-level image threshoding method, which effectively used the features of the selected pixels in the gray-level image, including coordinates, gray level and gradient. The trained SVM can binarize all pixels in the gray-level image to get a resultant binary image. Experimental results reveal that the proposed method can effectively binarize a gray-level image, even an uneven-lighting image. Based on the proposed binarization method, the research proposes a new multilevel image thresholding method, which recursively transform the gray-level image into the resultant image with the desired pixel levels. Experimental results reveal that the proposed method can categorize the pixels with the same gray level adaptively, and the resultant multilevel image is promising. The research also proposes the SVM-based handwritten Chinese character recognition method. It first extracts the features for an input handwritten character. The coarse classification using mean feature vectors then gets the candidate characters. Finally, the SVM is trained by the training handwritings of the candidates and then determines the category of the input handwriting. According to the experiments, the rate of recognizing handwritten Chinese characters can be increased to 98.31%.
摘要........................................................I
Abstract...................................................II
誌謝......................................................III
第一章 緒論.................................................1
1-1 研究背景與動機..........................................1
1-2 研究目的................................................2
1-3 論文架構................................................3
第二章 基礎理論與背景介紹...................................4
2-1 線性規劃Linear Programming..............................4
2-2 非線性規劃Nonlinear Programming........................14
2-2-1 等式限制 Equality Constraints........................15
2-2-2 積極集合法Active Set Method..........................16
2-3 支撐向量機.............................................25
2-3-1 支撐向量機概念.......................................25
2-3-2 Hard-Margin 支撐向量機...............................27
2-3-3 Soft-Margin 支撐向量機...............................31
2-3-4 核心函數(Kernel Function)............................34
2-3-5 支撐向量機訓練率評估.................................36
2-3-6 多類別支撐向量機.....................................37
2-3-7 解二次規劃...........................................39
2-4 影像二值化.............................................40
2-4-1 Otsu Method..........................................43
2-4-2 Minimum Cross Entropy Method.........................45
2-4-3 Tsallis Entropy Method...............................46
2-4-4 NiBlack Method.......................................47
2-4-5 Sauvola Method.......................................48
2-4-6 Bernsen Method.......................................49
2-4-7 Eikvil Method........................................50
2-5 影像多值化.............................................51
2-5-1 Fast Otsu’s Method..................................53
2-5-2 Particle Swarm Optimization Method...................56
2-5-3 Honey Bee Mating Optimization Method.................59
2-6 手寫中文字辨識.........................................63
第三章 以SVM為基礎之影像二值化法...........................65
3-1 影像前處理.............................................65
3-2 Support Vector Machine.................................68
3-3 以SVM為基礎之二值化演算法..............................69
第四章 以SVM為基礎之影像多階層化法.........................72
4-1 從灰階影像擷取兩階層的訓練點...........................72
4-2 以SVM為基礎之影像多階層化演算法........................76
第五章 以SVM為基礎之離線手寫中文字辨識.....................82
5-1 手寫中文字的特徵抽取...................................82
5-2 手寫中文字辨識演算法...................................87
第六章 實驗結果............................................90
6-1 以SVM為基礎之影像二值化法之實驗結果....................90
6-2 以SVM為基礎之影像多階層化法之實驗結果..................96
6-3 以SVM為基礎之離線手寫中文字辨識之實驗結果..............99
第七章 結論...............................................101
參考文獻..................................................102
[1]V.N. Vapnik, “The Nature of Statistical Learning Theory, Springer,“ New York,1995.
[2]Ying Yu , Yang Yang1 , “Similar Chinese Characters Recognition Using Multiresolution Feature Space ,“International Conference on IntelligentSystems Design and Application,2006.
[3] Fang Yang, Xue-Dong Tian, Xin Zhang , “An Improved Method for Similar Handwritten Chinese Character Recognition,” IEEE, 2010 , pp. 419-412
[4]G.Guo,S.Z.Li, and K.L. Chan, “Support Vector machines for face recognition,” Image and Vision Computing, Vol. 19, Aug 1,2001, pp.631-638
[5]B.Heisele,P.Ho, and T. Poggio, “Face recognition with support vector machines:Global versuscomponent-based approach,” International Conference on Computer Vision, Vol. 2, 2001, pp. 688-694
[6]Boser B. E., I. Guyon, and V. N. Vapnik, “A Training Algorithm for Optimal Margin Classifiers,” Annual Workshop on Computational Learning Theory, Vol. 5, 1992, pp. 144-152.
[7]Kuhn, H. W.; Tucker, A. W.., “Nonlinear programming. Proceedings of 2nd Berkeley Symposium,” Berkeley: University of California Press, 1951, pp. 481-492.
[8]R. Kohavi, 1995, “A study of cross-validation and bootstrap for accuracy estimation and model selection,” IJCAI, pp.1137-1145.
[9]Gass ,Saul I ,” Linear programming methods and applications”
[10]Computing in Science and Engineering, Vol. 2, No. 1, 2000
[11]D. G. Luenberger, “Linear and Nonlinear Programming,” Assison-Wesley, 1984.
[12]N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Trans. Systems Man Cybernetics, Vol. SMC-9, No. 1, 1979, pp. 62-66.
[13]J. Kittler, J. Illingworth, “Minimum error thresholding,” Pattern Recognition, Vol. 19, 1986, pp. 41-47.
[14] C.H. Li, C.K. Lee, “Minimum cross entropy thresholding,” Pattern Recognition, Vol. 26, No. 4, 1993, pp. 617-625.
[15] M. Portes de Albuquerque, I.A. Esquef, A.R. Gesualdi Mello, “Image thresholding using Tsallis entropy,” Pattern Recognition Letters, Vol. 25, 2004, pp. 1059-1065.
[16] S.D. Yanowitz, A.M. Brukstein, “A new method for image segmentation,” Comput. Vision Graphics Image Process, Vol.46, 1989, pp. 82-95.
[17] Ilya Blayvas, Alfred Bruckstein, Ron Kimmel, “Efficient computation of adaptive threshold surfaces for image binarization,” Pattern Recognition, Vol. 39, 2006, pp. 89-101.
[18] W. Niblack, “An introduction to digital image processing,” Prentice-Hall, Englewood Cliffs, NJ, 1986.
[19] L. Eikvil, T. Taxt, K. Moen, “A fast adaptive method for binarization of document images,” International Conference on Document Analysis and Recognition, France, 1991, pp. 435-443.
[20] J.R. Parker, C. Jennings, A.G. Salkauskas, “Thresholding using an illumination model,” ICDAR, Vol. 93, 1993, pp. 270-273.
[21]J. Sauvola, M. Pietikainen, “Adaptive document image binarization,” Pattern Recognition, Vol. 33, 2000, pp. 225-236.
[22]J. Yang, Y. Chen, W. Hsu, “Adaptive thresholding algorithm and its hardware implementation,” Pattern Recognition Letters, Vol. 15,No. 2, 1994 , pp. 141-150.
[23]J.Bernsen,“Dynamic thresholding of gray-level images,”Proc.Eigh-th Int'l Conj Pattern Recognition,Paris, 1986 ,pp. 1251-1255
[24]M. Chang, S. Kang, W. Rho, H. Kim, D. Kim, “Improved binarization algorithm for document image by histogram and edge detection,” ICDAR, Vol. 95, 1993, pp. 636-643.
[25]O.D. Trier, A.K. Jain, “Goal-directed evaluation of binarization methods,” IEEE Trans. Pattern Anal. Mach. Intell. , Vol. 17, No. 12, 1995, pp. 1191-1201.
[26]I.K. Kim, D.W. Jung, R.H. Park, “Document image binarization based on topographic analysis using a water flow model,” Pattern Recognition, Vol. 35, 2002, pp. 265-277.
[27]Qingming Huang , Wen Gao , Wenjian Cai, “Thresholding technique with adaptive window selection for uneven-lighting image,” Pattern Recognition Letters, Vol. 26, No. 6, 2005, pp. 801-808.
[28] http://140.134.32.129/scteach/matlabdsp/dsp2.htm.
[29] V.N. Vapnik, “An overview of statistical learning theory,” IEEE Trans. Neural Networks, Vol. 10, No. 5, 1999, pp. 988-999.
[30]Ethem Alpaydin, “Introduction to Machine Learning,” MIT Press, 2004.
[31]Y. Liu, Y.F Zheng “A novel feature selection method for support vector machines,” Pattern Recognition, Vol. 39, No. 7, 2006, pp. 1333-1345.
[32]F. Kurugollu, B. Sankur, A.E. Harmanci, “Color image segmentation using histogram multi-thresholding and fusion,” Image and Vision Computing, Vol. 19, 2001, pp. 915–928.
[33] V. Boskovitz, H. Guterman, “An adaptive neuro-fuzzy system for automatic image segmentation and edge detection”, IEEE Transactions on Fuzzy Systems, Vol. 19, No. 47, pp.–262.
[34] F. Du, W.K. Shi, L.Z. Chen, Y. Deng, A. Zhu, “Infrared image segmentation with 2-D maximum entropy method based on particle swarm optimization (PSO),” Pattern Recognition Letters, Vol. 26, 2005, pp. 597–603.
[35]C. Strouthopolos, N. Papamarkos, “Multi-thresholding of mixed-type documents,” Engineering Applications of Artificial Intelligence, Vol. 13, 2000, pp. 323–343.
[36] E. Azhara, S.K.S. Fan, D.M. Tsai, “Optimal multi-thesholding using a hybrid optimization approach,” Pattern Recognition Letter,Vol. 26, 2005, pp. 1082–1095.
[37] P.S. Liao, T.S. Chen, P.C. Chung, “A fast algorithm for multilevel thresholding,” Journal of Information Science and Engineering, Vol. 17, 2001, pp. 713–727.
[38] M. Madhubanti, A. Amitava, “A hybrid cooperative-comprehensive learning based algorithm for image segmentation using multilevel thresholding, “Expert Systems with Application,Vol. 34, 2008, pp. 1341–1350.
[39] P.Y. Yin, “Multilevel minimum cross entropy threshold selection based on particle swarm optimization,” Applied Mathematics and Computation, Vol. 184, 2007, pp. 503–513.
[40] M.H. Horng, “A multilevel image thresholding using the honey bee mating optimization,” Applied Mathematics and Computation, Vol. 215, 2010, pp. 3302–3310.
[41]S. Arora, J. Acharya, A. Verma, Prasanta K. Panigrahi, “Multilevel thresholding for image segmentation through a fast statistical recursive algorithm,” Pattern Recognition Letters, Vol. 29, 2008, pp. 119–125.
[42]J. Kennedy, R.C. Eberhart, “Particle swarm optimization,” in: Proc. IEEE Int’l. Conf. on Neural Networks, vol. IV, 1995, pp. 1942–1948.
[43]H.A. Abbasss, “Marriage in honey bee optimization (HBO): a haplometrosis polygynous swarming approach,” in: The Congress on Evolutionary Computation (CEC001), 2001, pp. 207–214.
[44]Mohammad Fathian, Babak Amiri *, Ali Maroosi, “ Application of honey-bee mating optimizationalgorithm on clustering,” Applied Mathematics and Computation , 2007, pp. 1502–1513.
[45]H.W. Hao, X.H. Xiao, and R.W. Dai, “Handwritten Chinese Character Recognition by Metasynthetic Approach,” Pattern Recognition,Vol. 30, No. 8, 1997, pp. 1321–1328.
[46]F.H. Cheng, “Multi-Stroke Relaxation Matching Method for Handwritten Chinese Character Recognition,” Pattern Recognition, Vol. 31, No. 4, 1998, pp. 401–410.
[47] T. Wakabayashi, Y. Deng, S. Tsuruoka, F. Kimura, and Y. Miyake,“Accuracy Improvement by Nonlinear Normalization and FeatureCompression in Handwritten Chinese Character Recognition,”Trans. IEICE, vol. J79-D-II, No. 5, May 1996, pp. 765–774.
[48] K. Saruta, N. Kato, M. Abe, and Y. Nemoto, “High Accuracy Recognition of ETL9B Using Exclusive Learning Neural Network—II (ELNET–II),” Trans. IEICE, vol. E79-D, No. 5, May 1996, pp. 516–522.
[49] Youbin Chen, Xiaoqing Ding and Youshou Wu., “ A new nonlinear shape normalization method for off-line handwritten Chinese character recognition, “ PR&AI, Vol. 11, No. 3, 1998, pp. 310-317.
[50]C. Pisarn, T. Theeramunkong, “An HMM-based method for Thai spelling speech recognition,” Computers and Mathematics with Applications ,Vol. 54 , 2007, pp.76–95.
[51]Ying Yu, C.L. Dong . “Similar Chinese characters recognition based on dynamical feature selection,” Computer Engineering, Vol. 32 , 2006 , pp.10-11.
[52] Fang Yang, Xue-Dong Tian, Xin Zhang, Xin-Biao Jia, “ An Improved Method for Similar Handwritten Chinese Character Recognition,” Third International Symposium on Intelligent Information Technology and Security Informatics, pp. 419-422.
[53]Ying Yu, Yang Yang,Cailin Dong,Xiuling He,Zengzhao Chen, “ Similar Chinese Characters Recognition Using Multiresolution Feature Space,” International Conference on Intelligent Systems Design and Applications,Vol 6, 2006.
[54]Cheng-huang Tung, Hsi-jian Lee, “Increasing character recognition accuracy by detection and correction of erroneously identified characters,” Pattern Recognition, Vol. 27, No. 9, pp. 1259-1266, 1994.
[55]Lixin Zhen, Ruiwei Dai. “Off-Line Handwritten Chinese Character Recognition with Nonlinear Pre-classification,” Proceedings of ICMI'2000, pp. 473~479, 2000.
[56]陳承昌,“支持向量機及Plausible Neural Network於水稻田辨識之研究”, 國立交通大學碩士土木工程學系研究所碩士論文,2006。
[57]張僩鈞,“兩階層式垃圾郵件過濾機制之研究”, 明傳大學資訊傳播工程學系碩士班碩士論文,2006。
[58]陳俊瑋,“一個擷取Gabor特徵的SVM人臉辨識方法”, 國立台灣科技大學電機工程研究所碩士論文,2006。
[59]林維謙,“植基於支援向量機之人臉偵測與人臉辨識”, 世新大學管理學院資訊管理研究所碩士論文,2007。
[60]陳煒燁,“利用SVM分類演算法於EEG-P300的偵測與分析”, 逢甲大學自動控制工程學系碩士班碩士論文,2008。
[61]蕭博文,“以SVM為基礎之偽鈔辨識”, 國立台灣科技大學電機工程研究所碩士論文,2009。
[62]謝芳伶,“支援向量機於指紋辨識之應用”,靜宜大學資訊管理學系碩士班碩士論文,2004。
[63]廖君瑋,“具區域筆順及比例結構變異容忍之線上手寫中文辨識”,國立屏東商業技術學院資訊科技應用研究所碩士論文,2008。
[64]工研院中文字形資料庫。
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top