簡易檢索 / 詳目顯示

回結果列表
研究生： 蘇義明
Su, Yih-Ming
論文名稱： 中文信封文字辨識與雜訊文字筆劃擷取
Recognition of Chinese Postal Addresses and Stroke Extraction from Noisy Characters
指導教授： 王駿發
Wang, Jhing-Fa
學位類別： 博士
Doctor
系所名稱： 電機資訊學院 - 資訊工程學系
Department of Computer Science and Information Engineering
論文出版年： 2003
畢業學年度： 91
語文別： 中文
論文頁數： 142
中文關鍵詞： 地址辨認 、筆劃擷取 、地址切割和字元辨認 、前處理 、影像擷取
外文關鍵詞： address recognition, stroke extraction, image acquisition, address segmentation, character recognition
相關次數： 點閱：98  下載：4 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 處理中文字的一些問題可從兩個觀點來考慮，其中包含以系統實作的觀點來完成地址辨認和以技術改善的觀點來完成筆劃擷取。這些考慮可提供以科學和工程的角度來研究中文字元的處理。首先，為了完成一套實作系統並考慮其實際的問題，我們發展一套郵件分揀系統，並根據中文字的處理以改善郵件處理的效率。這套系統以辨認中式信封上的縣市名和路街名，分別提供兩個不同的分揀階段，包含郵政中心和支局的分揀，而這套即時辨認系統組成幾個模組：影像擷取、前處理、地址切割和字元辨認等模組。在這些模組中，我們常用一些探索和嘗試錯誤的方法來解決一些實際的問題，包含處理低品質的影像、要求系統處理的速度和有效性以及系統操作的穩定性。從實驗觀察，我們完成第一階段縣市名的辨認，並以每小時分揀八千筆信封，其正確率為75.6%、錯誤率為0.92%和拒絕率為23.48%。再者，我們也完成第二階段路街名辨認，其正確率為71.6%、錯誤率為1.05%和拒絕率為27.35%。
    對於中文字處理在技術改善方面，我們也提出一個嶄新的筆劃擷取方法。以兩種方向性濾波的技術擷取雜訊中文字元的筆劃線段。對於使用結構分析作中文字辨認系統，此強健的擷取方法是必要的。此濾波技術分別使用兩個不同濾波核心，包含Gabor和SOGD函數。而且，為了擷取任意方向的筆劃線段，我們估算筆劃的方向是藉由發現筆劃線段在濾波輸出中有最大的能量反映。然後，我們利用學習式的特徵點確認方法，以減少字元結合點定位的不確定性，以取代特定性規則式的技術。從實驗結果觀察，我們所提出的筆劃線段擷取方法，不僅提供免除於結合點的失真和造成虛假的分支，而且容許擷取筆劃線段時，有雜訊存在於中文字元。再者，學習式的技術有助於提供一般性學習知識的能力對字元特徵點的瞭解且決定特徵點的型態，此方法優於規則式的技術對於字元特徵點的偵測。最後，我們使用SOGD濾波技術進一步分解一個中文字元進入不同方向的筆劃線段，同時我們也比較辨認的結果分別來自分解和非分解的字元處理，其中分解字元的處理有助於提高辨認率約17.31%。

    Some problems of processing Chinese characters are considered according to the aspects of the system implementation for address recognition and technique improvement for stroke extraction. The considerations give the scientific and engineering views for the research of character processing. Firstly, in order to take into account system implementation and practical issues, a mail-sorting system based on Chinese character processing is developed to highlight practical challenges for improving the effectiveness of automatic mail handling. The developed system is based on recognizing city/county and road/street names on the Chinese mail envelopes for two different sorting stages in mail sorting centers and branches, respectively. The proposed real-time system consists of the following modules: image acquisition, preprocessing, a new approach named curve-mapping segmentation, and neural network-based character recognition. Some heuristic approaches within these modules are adopted to improve some practical problems, including the processing of low-quality images, the demand of the processing speed, the robustness of system operation, and the effectiveness of system processing. From experimental observations, the first mail-sorting stage based on recognizing city/county names was achieved to sort 8000 mail pieces per hour with an accuracy rate of 75.6%, an error rate of 0.92%, and a rejection rate of 23.48%. Furthermore, the second mail-sorting stage based on recognizing road/street names was achieved in the error rate of 1.05%, the average accuracy rate of 71.6%, and the rejection rate of 27.35%.
    To consider technique improvement for Chinese character processing, we also proposed a novel stroke extraction method based on two directional filtering techniques for extracting reliable stroke segments of noisy and degraded Chinese characters. The robust stroke extraction method for Chinese characters is essential to off-line character recognition systems, which depend on stroke structure analysis to function. The filtering techniques use two different filter kernels, including Gabor and second-order Gaussian derivative (SOGD) functions. Moreover, to extract the stroke segments along an arbitrary orientation, the orientation of stroke segments is estimated by finding their maximal power responses in filter output. Then, a learning-based feature-point identification technique is used to eliminate the uncertainty of junction positioning, instead of using specific rule-based techniques for stroke-segment merging. From experimental observations, the proposed stroke-segment extraction approach based on the directional filtering techniques can not only provide immunity against the junction-distortion and spurious-branch problems caused from a thinning-based process, but is also insensitive to shape degradation and noise. Moreover, the learning-based technique is capable of generalizing the learning knowledge to determine the types of feature points, rather than the rule-based techniques for detecting feature points. Finally, the SOGD filtering-based technique is utilized to decompose the Chinese characters into a set of stroke segments. Meantime, comparing the recognition performance from the decomposition process with that from non-decomposition process reveals that the process of decomposing Chinese characters is very helpful for improving recognition performance about 17.31% in test character set.

    List of Figures List of Table 1. Introduction… … ..… … … .… … … … … … … … … … … … … … … … … … … … … … … .1 1.1 Motivations and Difficulties… … … … … … … … … … … … … ...… … … … … … … 1 1.2 Research Directions and Survey… … … … … … … … … … … ..… … … … … … … ...2 1.2.1 Dissertation Study Directions… … … … … … … … … … … … … … … … … … 2 1.2.2 Survey of Mail Sorting System… … … … … … … … . … … … … … … … … ...4 1.2.3 Survey of Stroke Extraction… … … … … … … … … … … … … … … … … … ..5 1.3 Overview of Proposed Approaches… … … … … … … … … ..… … … … ..… .… … ...6 1.3.1 City/County-Name Processing by Character Recognition… … … … … … … 7 1.3.2 Road/Street-Name Processing by Word Recognition… … … … … … .… … ...7 1.3.3 Stroke-Segment Extraction by Gabor Filtering… … … … … … … … … … … .8 1.3.4 Stroke-Segment Extraction by SOGD Filtering… … … … … … … … … … … 8 1.3.5 Stroke Merging by Feature-Point Learning… … … … … … … … … … … … ...9 1.4 Organization of the Dissertation… … … … … … … … … … … … … … … … … … ...10 2. Recognition of City/County Names… … … … … … … ..… ...… … … … … … … … ..11 2.1 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … ..11 2.2 Interlaced Image Acquisition… … … … … … … … … … … … … … … … … … … … 12 2.3 Preprocessing and Address Block Location… … … … … … … … … … … … … … .13 2.3.1 Merging of even and odd images… … … … … … … … … … … … … … … … 13 2.3.2 Skew Correction and Translation … … … … .… … … … … … … … … … … ..14 2.3.3 Image Binarization… … … … … … … … … … … … … … … … … … ....… … ..14 2.3.4 Address Block Location… … … … … … … … … … … … … … … … … … … ..15 2.4 Segmentation and Extraction of Characters… … … … … … … … … … … .… … … 17 2.4.1 Coarse Segmentation… … … … … … … … .… … … … … … … … … … … … ..17 2.4.2 Parameter Estimation… … … .… … … … ..… … … … … … … … … … … … ...18 2.4.3 Fine Segmentation… … … … .… … … … … … … .… … … … … … … … … … 19 2.4.4 Segment Combination… … … … … … … … … … … … … … … … … … … … .21 2.5 Character Recognition… … … … … … … … … … … … … … … … … … … … … … ..22 2.5.1 Meshing-Based Feature Extraction… … … … … … … ..… … … … … … … ...23 2.5.2 Learning-Based Character classification… … … … ...… … … … … … … ..… 24 2.5.3 Contextual Postprocessing… … … … … … … … … … … … … … … … … … ..24 2.6 Experimental Results and Discussions… … … .… … … … … … … … … … … … … 25 2.6.1 Performance Analysis… … … … … … ..… … … … … … … … … … … … … ....26 2.6.2 Discussions… … … … .… … … … … … … … .… … … … … … … … … … … … 30 2.7 Concluding Remarks… … … … … … … … … … … … … … … … … … … … … … … 30 3. Recognition of Road/Street Names… … … … … … ..… … … ...… ...… … … ..… … ..32 3.1 Introduction… … … … … … … … … … … … … … … … … … … … ..… … … … … … 32 3.2 Pre-Classification with Fuzzy Rule-Based Approach… … … … … … … … … … ..33 3.2.1 Feature Extraction using Character Complexity… … … … … … … … … .… 35 3.2.2 Training Phase by K-Means Clustering. … … … … … … … … … … … … … 35 3.2.3 Classification Phase by Fuzzy Decision Approach… … … … … … … … … .36 3.3 Fine Classification with HMM-Based Approach… … … … … … … … … .… … … 38 3.3.1 Meshing-Based Feature Extraction… … … … … … … ..… … … … … .… … ..40 3.3.2 Training Phase… … … … .… … … … … ..… … … … … … … … … … … … … ..40 3.3.3 Recognition phase… … … … … … … … … … … … … … … … … … … … … ...42 3.4 Experimental Results and Discussions… … … .… … … … … … … … … … … … … 43 3.4.1 Performance Analysis… … … … … … ...… … … … … … … … … … … … .… ..43 3.4.2 Discussions… … … … .… … … … … … .… … … … … … … … … … … … … ....46 3.5 Concluding Remarks… … … … … … … … … … … … … … … … … … … … … … … 47 4. Gabor Filtering for Extracting Stroke Segments… … … … … … … … … … ....… .49 4.1 Introduction… … … … … … … … … … … … … … … … … … ..… … … … … … … … 49 4.2 Extracting Directional Features… … … … … … … … … … … … ..… … … … … … ..49 4.2.1 Design of 1-D Gabor Filters… … … … … … .… … … … … … … … … … … ...50 4.2.2 Parameter Selection… … … … … … … … … ..… … … … … … … … … … … ...52 4.3 Threshold Selection… … … … … … ...… … … … … … … … … … … … … … … … ...53 4.4 Refinement Process… … … … … … … … … … … … … … … … … … … … .… … … .55 4.5 Merging Process… … … .… … … … … … … … … … … … … … … … … … .… … … .58 4.6 Experimental Results and Discussions… … … .… … … … … … … .… . … … … … .61 4.6.1 Performance Analysis… … … … … … . … … … … … … … … … … … .… … ...61 4.6.2 Discussions… … … … .… … … … … … … … … … … … … … … … … … … … .62 4.7 Concluding Remarks… … … … … … … … … … … … … … … … … … … … … … … 63 5. SOGD Filtering for Extracting Stroke Segments… … … … … … … … … … ...… .70 5.1 Introduction… … … … … … … … … … … … … … … … … … … … … … … … … … ..70 5.2 Extracting Directional Features… … … … … … … … … … … … … … … … … … … 71 5.2.1 Design of SOGD Filters.… ......… … … … … .… … … … … … … … … … … ..71 5.2.2 Parameter selection… … … … … … … … … ..… … … … … … … … … … ....… 73 5.3 Extracting Stroke Segments along an Arbitrary Orientation… … … … … … ..… ..74 5.4 Refinement and Vectorization Processes… … … … … … … … … … … … … … … ..77 5.5 Character Classification… … … … … … … … … … … … … … … … … … … … … ...78 5.5.1 Structural Features.… … … … … … … … … … … … … ...… … … … … … … ...79 5.5.2 Statistical Features… … … … … … … … … ..… … … … … … … ...… … … … ..80 5.5.3 Classification Processing… … … … … … … ..… … … … … ...… … … … … … 81 5.6 Experimental Results and Discussions… … … .… … … … … … … … … .… . … … .82 5.6.1 Performance Analysis… … … … … … … … … … … … … … … … … … ..… … 82 5.6.2 Discussions… … … … .… … … … … . … … … … … … … … … … … … … … ...89 5.7 Concluding Remarks… … … … … … … … … … ...… … … … … … .… … … … … … 91 6. Learning-based Feature-Point Identification… … ...… … … … … ..… . … … … … 92 6.1 Introduction… … … … … … … … … … … … … … … … … ...… … … … .… … … … ..92 6.2 Finding Associated Endpoints...… … … … … … … … … … ...… … … … .… … … ...94 6.3 Modeling Feature Points… … … … … … … … … … … … … … ..… .… .… … … … ..96 6.4 Merging Feature Points into a Junction Point… … … … … … … ...… .… ....… … ..99 6.5 Experimental Results and Discussions… … … .… … … … … … … … .… … … … .101 6.5.1 Performance Analysis… … … … … … … ..… … … … … … … … … … … … ..102 6.5.2 Discussions… … … … .… … … … … … ..… ...… … … … … … … .… … … … .104 6.6 Concluding Remarks… … … … … … … … … ...… … … … … … … … … .… … … ..105 7. Conclusions… ...… … … … … … ...… … … … … … … ..… ...… … … … .....… … .… … ...109 BIBLIOGRAPHY… … … … ..… … ...… … … … … … … … .… … … … … … … ..… … .… 112 PUBLICATION LIST… … … … ...… … … … … … … … .… … … … … … … … … … .… .122

    [1]. W. C. Kao and T. M. Parng, “Integrating statistical and structural approaches to handprinted Chinese character recognition,” IEICE Trans Inf. and Syst., vol. E81-D, no. 4, pp.391-400, 1998.
    [2]. F. H. Cheng, “Multi-stroke relaxation matching method for handwritten Chinese character Recognition,” Pattern Recognition, vol. 31, no. 4, pp. 401-410, 1998.
    [3]. C. W. Lee and Z. Chen, “Handwritten Chinese character recognition using stroke structural sequence code,” J. of Information Science and Engineering, vol. 13, pp. 631-648, 1997.
    [4]. A. J. Hsieh, K. C. Fan, T. I. Fan, and C. W. Ho, “A stroke-based handwritten Chinese character recognition system using greedy matching method,” J. of Information Science and Engineering, vol. 11, pp. 1-22, 1995.
    [5]. F. H. Chang, W. H. Hsu, and M. C. Kuo, “ Recognition of handprinted Chinese characters via stroke relaxation,” Pattern Recognition, vol. 26, no. 4, pp. 579-593, 1993.
    [6]. S. W. Lu, Y. Ren and C. Y. Suen, “Hierarchical attributed graph representation and recognition of handwritten Chinese characters,” Pattern Recognition, vol. 24, no. 7, pp. 617-632, 1991.
    [7]. L. H. Chen and J. R. Lieh, “Handwritten character recognition using an 2-layer random graph model by relaxation matching,” Pattern Recognition, vol. 23, no. 11, pp. 1189-1205, 1990.
    [8]. F. H. Cheng, W. H. Hsu, and M. Y. Chen, “Recognition of handwritten Chinese characters by modified Hough transform techniques,” IEEE Trans. On Pattern Analysis and Machine Intelligence, vol. 11, no. 4, 1989.
    [9]. H. D. Chang and J. F. Wang, “A robust stroke extraction method for handwritten Chinese characters,” Inter. J. of Pattern Recognition and Artificial Intelligence, vol. 8, no. 5, pp. 1223-1239, 1994.
    [10]. H. J. Lee and B. Chen, “Recognition of handwritten Chinese characters via short line segments,” Pattern Recognition, vol. 25, no. 5, pp. 543-552, 1992.
    [11]. Chin-Chuan, Han; Yao-Lung, Tseng; Kuo-Chin, Fan; An-Bang, Wang, “Coarse classification of Chinese characters via stroke clustering method,” Pattern Recognition Letters, vol. 16, pp. 1079-1089, 1995.
    [12]. A. B. Wang, K. C. Fan and J. S. Huang, “Recognition of handwritten Chinese characters by modified relaxation methods,” Image and Vision Computing, vol. 12, no. 8, pp. 509-522, 1994.
    [13]. Xue Gao, Lian-Wen Jin, Jun-Xun Yin, and Jian-Cheng Huang, “A new stroke-based directional feature extraction approach for handwritten Chinese character recognition,” Inter. Conf. On Document Analysis and Recognition, vol. 2, pp. 635-639, 2001.
    [14]. X. M. Sun, “A new stroke extraction method of Chinese character,” Inter. J. of Pattern Recognition and Artificial Intelligence, vol. 5 no. 4, pp. 707-721, 2001.
    [15]. K. C. Fan and W. H. Wu, “A run-length-coding-based approach to stroke extraction of Chinese characters,” Pattern Recognition, vol. 33. pp. 1881-1895, 2000.
    [16]. H. P. Chiu and D. C. Tseng, “A novel stroke-based feature extraction for handwritten Chinese character recognition,” Pattern Recognition, vol. 32, pp. 1947-1959, 1999.
    [17]. J. Y. Lin and Z. Chen, “A Chinese-character thinning based on global and features and contour information,” Pattern Recognition, vol. 28, no. 4, pp. 493-512, 1995.
    [18]. L. Y. Tseng and C. T. Chuang, “An effect knowledge-based stroke extraction method for multi-font Chinese characters,” Pattern Recognition, vol. 25, no. 12, pp. 1445-1458, 1992.
    [19]. C. T. Chuang and L. Y. Tseng, “A stroke extraction method for multi-font Chinese characters based on the reduced special interval graph,” IEEE Trans. On Systems, Man, Cybernetics, vol. 25, no. 7, pp. 1171-1178, 1995.
    [20]. Y. S. Chen, “Segmentation and association among lines and junctions for line image,” Pattern Recognition, vol. 27, no. 9, pp. 1135-1157, 1994.
    [21]. C. N. Lee and B. H. Wu, “A Chinese-character-stroke-extraction algorithm based on contour information,” Pattern Recognition, vol. 31, no. 6, pp. 651-663, 1998.
    [22]. R. He and H. Yan, “Stroke extraction as pre-processing step to improve thinning results of Chinese characters,” Pattern Recognition Letter, vol. 21, pp. 817-825, 2000.
    [23]. K. M. Ku and P. P. K. Chiu, “Fast stroke extraction method for handwritten Chinese characters by cross region analysis,” Electronics Letters, vol. 30, No. 15, pp. 1210-11212, 1994.
    [24]. K. Liu, Y. S. Huang, and C. Y. Suen, “Identification of fork points on the skeletons of handwritten Chinese characters,” IEEE Trans. Pattern Analysis and Machine Intelligence, 21, 10 (1999) 1095-1100.
    [25]. H. Ogawa, and K. Taniguchi, “Thinning and stroke segmentation for handwritten Chinese character recognition,” Pattern Recognition, vol. 15, no. 4, pp. 299-308, 1982.
    [26]. J. R. Lin and C. F. Chen, “Stroke extraction for Chinese characters using a trend-followed transcribing technique,” Pattern Recognition, vol. 29, no. 11, pp. 1789-1805,1996.
    [27]. C. W. Liao, and J. S. Huang, “Stroke segmentation by Bernstein-Bezier curve fitting,” Pattern Recognition, vol. 23, no. 5, pp. 475-484, 1990.
    [28]. A. Amin and S. Singh, “Recognition of hand-printed Chinese characters using decision tree machine learning system,” Pattern Analysis and Application, vol. 1, pp. 130-140, 1998.
    [29]. J. W. Kim, I. K. Kim, B. J. Choi, and H. J. Kim, “Decomposition of Chinese character into strokes using mathematical Morphology,” Pattern Recognition Letters, vol. 20, pp. 285-292, 1999.
    [30]. T. Pavlidis, “A vectorizer and feature extractor for document,” Computer Vision, Graphics, and Image Processing, vol. 35, pp111-127, 1986.
    [31]. J. A. Nelder and R. Mead, “A simplex method for function minimization,” Comput. J., vol. 7, pp.308-313, 1965.
    [32]. C. G. Loonly, “Pattern recognition using neural networks- theory and algorithms for engineers and scientists,” New York, Oxford University Press. Inc, 1997.
    [33]. L. Lam, Y.S. Huang, and C. Y. Suen, “Combination of multiple classifier decisions for optical character recognition,” Handbook of Character Recognition and Document Image Analysis, World Scientific Publishing Co., pp. 79-101, 1997.
    [34]. J. Daugman, “Uncertainty relation for resolution in space, spatial frequence, and orientation, optimized by two-dimensional visual cortical filters, J. Opt. Soc. Am., vol. 2, no. 7, pp.1160-1169, 1985.
    [35]. O. Nestares, R. Navarro, J. Portilla, and A. Tabernero, “Efficient spatial-domain implementation of a multi-scale image representation based on Gabor functions,” J. Electronic Imaging, vol. 7, pp. 166-173,1998.
    [36]. R.C. Gonzalez and R.E. Woods, “Digital Image Processing,” Addison-Wesley, Publish Company, Inc., Reading Massachusetts, pp. 218-221, 1992.
    [37]. R. Buse, Z. Q. Liu, and T. Caell, “Using Gabor filters to measure the physical parameters of line,” Pattern Recognition,” vol. 29, no. 4, pp. 615-625, 1996.
    [38]. M. W. Carter, “The indefinite zero-one quadratic problem,” Discrete Applied Mathematics, vol. 7, pp. 23-44,1984.
    [39]. R. M. Haralick and L.G. Shapiro, “Computer and Robot Vision,” Addision-Wesley Publishing Co., pp.62-73, 1992.
    [40]. J. J. Koenderink, “Operational significance of receptive field assemblies,” Biological Cybernetics, vol. 58, pp. 163-171, 1988.
    [41]. N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Trans. Systems, Man, and Cybernetics, vol. 9. no. 1 pp. 377-393,1979.
    [42]. T. C. Folsom and R. B. Pinter, “Primitive features by steering quadrature, and scale,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 20, no. 11, pp. 1161-1173, 1998.
    [43]. L. Lam, S. W. Lee and C. Y. Suen, “Thinning methodologies- A comprehensive survey,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 14 no. 9, pp. 869-885, 1992.
    [44]. Serge Lang and Gene Murrow, “Geometry: a high school course,” Spring-Verlag New York Inc. 1988.
    [45]. K. W. Seo, J. Y. Kim, J. M. Yoon, and K. S. Chung, “Comparison of feature performance and its application to feature combination in off-line handwritten Korean Alphabet recognition,” International Journal Pattern Recognition and Article Intelligence, vol. 12, no. 2, pp. 251-261, 1998.
    [46]. F. Kimura, T. Wakabayashi, S. Tsuruoka, and Y. Miyake, “Improvement of handwritten Japanese character recognition using weighted direction code histogram,” Pattern Recognition, vol. 30, No. 8, pp. 1329-1337, 1997.
    [47]. K. Fukunaga, “Introduction to statistical pattern recognition,” Academic Press, Inc. New York, Second Edition, 1990.
    [48]. H. J. Trussel, “Picture thresholding using an iterative selection method,” IEEE Trans. System, Man, and Cybernetics, vol. 9, pp. 311-319, 1979.
    [49]. J. R. Parker, “Algorithms for image processing and computer vision,” John Wiley and Sons Inc., New York, pp.116-145, 1996.
    [50]. Y. Y. Zhang and C.Y. Suen, “A fast parallel algorithm for thinning digital pattern,” Communication of the ACM, vol. 27, no. 3, pp.236-239, 1985.
    [51]. R. Kohavi, “A study of cross-validation and bootstrap of accuracy estimation and model selection,” Proc. International Joint conference on Artifical Intelligence, pp. 1137-1143, 1995.
    [52]. Y. M. Su and J. F. Wang, “A novel stroke extraction method for Chinese characters using Gabor filters”, Pattern Recognition, vol. 36, no. 3, pp. 635-647, 2003.
    [53]. F. Sun, S. Omachi, and H. Aso, “Precise selection of candidates for handwritten character recognition using feature regions,” IEICE Trans. Inf. & Syst., Vol. E79-D, no. 5, pp.510-515, 1996.
    [54]. R. Buse, Z.Q. Liu and T. Caelli, “A structural and relational approach to handwritten word recognition,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 27, no. 5, pp. 847-861, 1997.
    [55]. Y. Hamamoto, S. Uchimura, M. Watanabe, T. Yasuda, Y. Mitani, and S. Tomita, “A Gabor filter-based method for recognizing handwritten numerals,” Pattern Recognition, vol. 31, No. 4, pp. 395-400, 1998.
    [56]. N. Srihari, “High-performance reading machines,” Proc. of the IEEE, 80, 1120-1132, 1992.
    [57]. S. N. Srihari, “Recognition of handwritten and machine-printed text for postal address interpretation,” Pattern Recognition letters, 14, 291-302, 1993.
    [58]. V. Govindaraju, A. Shekhawat and S.N. Srihari, “Interpretation of handwritten addresses in US mail-stream,” The Third Inter. Workshop on Frontiers in Handwriting Recognition, 197-206, 1993.
    [59]. P. W. Palumbo and S. N. Srihari, “Postal address reading in real time,” Inter. J. of Imaging Science and Technology, 7, 370-378, 1996.
    [60]. S. N. Srihari and E. J. Kuebert, “Integration of hand-written address interpretation technology in the United States postal service remote computer reader system,” Inter. Conf. Document Analysis and Recognition, 2, 892-896, 1997.
    [61]. A. C. Downton, R. W. S. Tregidgo and C.G. Leedham, and Hendrawan, “Recognition of handwritten British postal addresses,” From Pixels to Features III: Frontiers in Handwriting Recognition, 129-143, 1992.
    [62]. A. Cuhadar and A. C. Downton, “Structured parallel design for embedded vision system: an application case study,” Inter. Conf. On Image Processing and its Applications, 712-716, 1995.
    [63]. Y. Tokunaga, "History and current state of postal mechanization in Japan," Pattern Recognition Letters, 14, 4, 277-280, 1993.
    [64]. N. Nakajima, T. Tsuchiya, T. Kamimura, K. Yamada, “Analysis of address layout on Japanese handwritten mail- a hierarchical process of hypothesis verification,” Proc. of the 13th Inter. Conf. on Pattern Recognition, 726-731, 1996.
    [65]. S. W. Lee and K. C. Kim, “Address block location on handwritten Korean envelopes by the merging and splitting method,” Pattern Recognition, 27, 1641-16511, 1994.
    [66]. S. W. Lee and E. S. Kim, “Efficient postprocessing algorithms for error correction in handwritten Hangul address and human name recognition,” Pattern Recognition, 27, 12, 1631-1640, 1994.
    [67]. C. G. Leedham and P. E. Jones, “Automatic sorting of Australian handwritten letter mail using OCR and address feature verification,” TENCON’92, 287-291, 1992.
    [68]. Thmomas H. Hildebrandt and Wenta Liu, “Optical recognition of handwritten Chinese characters advance since 1980,” Pattern Recognition, 26, 2, 205-225, 1993.
    [69]. Shunji Mori, C. H. Suen and Kazuhiko Yamaoto, “Historical review of OCR research and development,” Proc. IEEE, 80, 7, 1029-1057, 1992.
    [70]. F. H. Cheng and W. H. Hsu, “Research on Chinese OCR in Taiwan,” Int. J. of Pattern Recognition and Artificial Intelligence, 5, 1&2, 139-164, 1991.
    [71]. H. Bunke and P. S. P. Wang, “Handbook of Character Recognition and Document Image Analysis,” Word Scientific Publishing Co., Singapore, pp. 331-355, 1997.
    [72]. Y. M. Su, C. L. Chen and J. F. Wang, “Comparative study of feature sets used in the recognition of Chinese handwritten characters,” International Symposium on Artificial Neural Networks, Taiwan, R.O.C., pp. D3-13-18, 1995.
    [73]. J. T. Ton and R. C. Gonzalez, “Pattern Recognition Principles,” Addison-Wesley Publishing Co., Massachusetts, pp. 94-97,1974.
    [74]. Valluru Rao and Hayagriva Rao, “C++ Neural Networks and Fuzzy Logic,” New York, Henry Holt and Company, INC., 1995.
    [75]. Z. Chi, H. Yan, and T. Pham, “Fuzzy Algorithms: With Application to Image Processing and Pattern Recognition,” Singapore, Word Scientific Publishing Co. Pte. Ltd. pp.139-187, 1996.
    [76]. Lawrence Rabiner and Biing-Hwang Juang, “Fundamentals of Speech Recognition,” New Jersey, Prentice-Hall, Inc., 1993.
    [77]. Y. M. Su and J. F. Wang, “Recognition of handwritten Chinese postal address using neural networks,” Proceedings - IEEE International Symposium on Circuits and Systems, 3, 25-28, 1998.
    [78]. Y. Mizukami, “A handwritten Chinese character recognition system using hierarchical displacement extraction based on directional features,” Pattern Recognition Letters, vol. 19, pp. 595-604, 1998.
    [79]. Lianwen Jin, Kwokping Chan, and Bingzheng Xu, “Off-line Chinese handwriting recognition using multi-stage neural network Architecture,” IEEE International Conference on Neural Networks, vol. 6, pp. 3083-3088, 1996.
    [80]. H. C. Fu and K. P. Chiang, “Recognition of handwritten Chinese characters by multi-stage neural network classifiers,” IEEE International Conference on Neural Networks, vol. 1, pp. 1303-1307, 1995.
    [81]. J. C. Bezdek and S. K. Pal, “Fuzzy Models for Pattern Recognition,” IEEE Press, New York, 1992.
    [82]. C. C. Lee, “Fuzzy logic in control systems: fussy logic controller- part II,” IEEE Trans. On Systems, Man, and Cybernetics, vol. 20, no. 2, pp.419-435, 1990.
    [83]. J. Makhoul, R. Schwartz, C. Lapre, and I. Bazzi, “A script-independent methodology for optical character recognition”, Pattern Recognition, vol. 31, no. 9, pp.1285-1294, 1998.
    [84]. D. Guillevic and C. Y. Suen, “HMM word recognition engine,” Proceedings of the Fourth International Conference on Document Analysis and Recognition, vol. 2, pp. 544-547, 1997.
    [85]. M. Mohamed and P. Gader, “Handwritten word recognition using segmentation-free hidden Markov modeling and segmentation-based dynamic programming techniques,” IEEE Trans. Pattern. Anal. Mach. Int., vol. 18, no. 5, pp. 548-554, 1996.
    [86]. H. Bunke, M. Roth, and E. G. Schukat-Talamazzini, “ Off-line cursive handwriting recognition using hidden Markov models,” Pattern Recognition, Vol. 28, No. 9, pp.1399-1413, 1995.

    下載圖示 校內：立即公開
    校外：2003-06-24公開
    QR CODE