本論文之目的在於為公文文件處理中的文件傾斜角度校正、資料欄位的擷取及數字辨識等問題，提出解決方法。在文件傾斜角度校正方面，利用公文文件大多是以文字組成的特性，以vertical run-length smoothing演算法處理影像後，再計算文字區塊的orientation，進而得到整張文件的傾斜角度。在資料欄位擷取方面，先以dilation及erosion二個morphological operations求得公文戳印的所在位置。之後，以特徵點群聚來分離戳印裡的表格格線與字元，再利用公文文號的字元特性及位置資訊，擷取出公文文號。而在數字辨識方面，以字元輪廓的傅利葉特徵搭配NNC分辨器，做為分辨數字的依據。以70張掃瞄解析器為300dpi的公文影像實驗之測試樣本，實驗結果顯示本研究之方法可以正確地擷取公文戳印及文號，且對於文號之辨識率可達98.7%；另外，對於NIST資料庫之手寫數字辨識率為90.39%。

This thesis addresses the problem of the skew detection, data extraction and digit recognition in the document analysis systems. Vertical run-length smoothing algorithm was applied to determine the skewness of the document and to adjust the image accordingly. Morphological operations, dilation and erosion, were used to locate the serial number field. Feature point clustering technique was used to separate digits from non-digits. The final recognition was carried out by a NNC using Fourier features obtained from digit contour information. Experiment on 700 characters obtained from 70 document images showed a 98.7% recognition rate. Furthermore, when applied to 10000 characters from NIST special database 19, 90.39% recognition rate was achieved.

[1] 張貴雲, 表單手寫欄位資料之萃取, 國立台灣師範大學資訊教育研究所碩士論文, 1998.
[2] 林育慈, 離線手寫印刷體英數字之辨識, 國立台灣師範大學資訊教育研究所碩士論文, 1998.
[3] Alessandro L. Koerich and Luan Ling Lee, “A system for automatic extraction of the user-entered data from bankchecks,” Proc. Int’l Symposium on Computer Graphics, Image Processing and Vision, SIBGRAPI’98, pp. 270-277, 1998.
[4] Alireza Khotanzad and Yaw Hau Hong, “Invariant image recognition by Zernike moment,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 12, no. 5, pp. 489-497, May 1990.
[5] Arnaud Ribert, Yves Lecourtier and Abdel Ennaji, “Designing efficient distributed neural classifiers: application to handwritten digit recognition,” Proc. Fifth Int’l Conf. Document Analysis and Recognition, ICDAR '99, pp. 265-268, 1999.
[6] Avanindra and Subhasis Chaudhuri, “Robust detection of skew in document images,” IEEE Trans. on Image Processing, vol. 6, no. 2, February 1997.
[7] Cha-Sup Jeong and Dong-Seok Jeong, “Hand-written digit recognition using Fourier descriptors and contour information,” Proc. of the IEEE Region 10 Conf. vol. 2, pp. 1283-1286, 1999.
[8] Dahai Cheng and Hong Yan, “Recognition of handwritten digits based on contour information,” Pattern Recognition, vol. 31, no. 3, pp. 235-255, 1998.
[9] Daniel S. Le, George R. Thoma and Harry Wechsler, “Automated page orientation and skew angle detection for binary document images,” Pattern Recognition, vol. 27, no. 10, pp. 1325-1344, 1994.
[10] F. Y. Shih, S.-S. Chen, “Adaptive Document Block Segmentation and Classification,” IEEE Trans. on Systems, Man, and Cybernetics - Part B, vol. 26, no. 5, pp. 797-802, October 1996.
[11] H. Yan, “Skew correction of document image using interline cross-correlation,” CVGIP: Graphical Models and Image Processing, vol. 55, pp. 538-534, 1993.
[12] James R. Parker, Algorithms for image processing and computer vision, 1996.
[13] Jing Wu and Hong Yu, “Combined SOM and LVQ based classifiers for handwritten digit recognition,” Proc. IEEE Int’l Conf. Neural Networks, vol. 6, pp. 3074-3077, 1995.
[14] Jinhai Cai and Zhi-Qiang Liu, “Integration of structural and statistical information for unconstrained handwritten numeral recognition,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 21, no. 3, March 1999.
[15] Kuo-Chin Fan, Jeng-Ming Lu, Liang-Sheng Wang and Hong-Yuan Liao, “Extraction of characters from form documents by feature point clustering,” Pattern Recognition Letters, vol. 16, pp. 963-970, 1995.
[16] Lawrence O’Gorman, “The document spectrum for page layout analysis,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 15, no. 11, November 1993.
[17] L. A. Torres-Méndez, J. C. Ruiz-Suárez, Luis E. Sucar and G. Gómez, “Translation, Rotation, and Scale-Invariant Object Recognition,” IEEE Trans. on Systems, Man, and Cybernetics, vol. 30, no. 1, February 2000.
[18] Minoru Fukumi, Sigeru Omatu and Yoshikazu Nishikawa, “Rotation-invariant neural pattern recognition system estimating a rotation angle,” IEEE Trans. on Neural Network, vol. 8, no. 3, pp. 568-581, May 1997.
[19] M. W. Chen and M. H. Ng, “Recognition of unconstrained handwritten numerals using crossing features,” Fifth Int’l Symposium on Signal Processing and its Application, ISSPA’99, Brisbane, Australia, vol. 1, pp. 22-25, August 1999.
[20] N. W. Stranthy and C. Y. Suen, “A new system for reading handwritten zip codes,” Proc. Third Int’l Conf. Document Analysis and Recognition, (ICDAR’95), vol. 1, pp. 74-77, 1995.
[21] R. Jain, R. Kasturi and B. B. Schunck, Machine Vision, 1995.
[22] Robert R. Bailey and Mandyam Srinath, “Orthogonal moment features for use with parametric and non-parametric classifier,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 18, no. 4, pp. 389-399, April 1996.
[23] S. Liang, M. Ahmadi, M. Shridhard, “Segmentation of touching characters in printed document,” Proceedings of the Second International Conference on Document Analysis and Recognition, pp. 569-572, 1993.
[24] S. Madhvanath, G. Kim and V. Govindaraju, “Chaincode contour processing for handwritten word recognition,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 21, no. 9, pp. 928-932, September 1999.
[25] S. N. Srihari and F. J. Keubert, “Integration of handwritten address interpretation technology into the United States postal service remote computer reader system,” Proc. Fourth Int’l Conf. Document Analysis and Recognition, Ulm, Germany, vol. 2, pp. 892-896, August 1997.
[26] Salim Djeziri, Fathallah Noubound and Rejean Plamondon, “Extraction of signatures from check background on a filiformity criterion,” IEEE Transactions on Image Processing, vol. 7, no. 10, pp. 1425-1438, October 1998.
[27] Seong-Whan Lee and Hee-Heon Song, “A new recurrent neural-network architecture for visual pattern recognitioin,” IEEE Trans. on Neural Network, vol. 8, no. 2, pp. 331-340, March 1997.
[28] Sung-Bae Cho, “Neural-network classifiers for recognizing totally unconstrained handwritten numerals,” IEEE Trans. on Neural Network, vol. 8, no. 1, pp. 43-53, 1997.
[29] T. Akiyama and N. Hagita, “Automatic entry system for printed documents,” Pattern Recognition, vol. 23, pp. 1141-1154, 1990.
[30] U. Pal, B. B. Chaudhuri, “An improved document skew angle estimation technique,” Pattern Recognition Letters, pp. 899-904, 1996.
[31] Xingyuan Li, Jiarong Hong, Zhaohui Zhang and Bin Chen, “A statistical form reading system,” 1993 IEEE Region 10 Conference on Computer, Communication, Control and Power Engineering, TENCON '93, vol. 2, pp. 1062-1065, 1993.
[32] Y. Y. Chung and M. T. Wong, “High accuracy handwritten character recognition system using contour sequence moments,” Proc. Fourth Int’l Conf. Signal Processing, vol. 2, pp. 1249-1252, 1998.