利用部分資訊做資料檢索，是一個常遇到的課題；而應用於表格文件時，往往需要較完整的表格文件資訊做比對檢索，若當表格文件留下部分資訊或文件中有特定的表格結構型態時，運用僅有的資訊透過檢索系統將相似的完整表格文件於系統資料庫作輸出，讓表格文件檢索能更具強健性與實用性。
本論文之研究目的為開發可辨識完整或部分表格且可容忍表格產生數位誤差(表格線遺失或增加)的檢索系統；利用邊角特徵編碼與區域延伸特徵，並結合於動態規劃法找出表格間最佳區域排比；本論文設計一相似度運算法則(Code-ratio-adjacent similarity, )，估算此區域排比相對於表格間的相似度值，以有效檢索候選表格於表格資料庫。進一步將候選表格做相對位置的定位，以完成表格文件檢索。
實驗結果顯示， 具備整合性特徵的相似度比對與運算法則，不僅只利用部分表格建立有效特徵資訊，又或者在數位誤差發生的情況下，仍可達成完整表格資料庫檢索與定位。

Using partial information retrieve that is a subject often encountered, and applies to form document, it often needs complete information of form document to retrieve. When form document remains partial information or has a specific structure of form type, our research can output the similar form document through the retrieval system.
In this paper, we propose a retrieval system which can recognize the partial form document and tolerate the digital error which is the lines decrease or increase of the form. Use the coding of corner pattern and extended corner feature to combine with dynamic programming method to find the optimal local alignment between the two forms. We design a similarity algorithm (Code-ratio-adjacent similarity) to estimate the similarity of optimal local alignment and search the candidate forms in the database. Further use the candidate forms to locate position with the partial form.
The experimental result showed Code-ratio-adjacent similarity was a similarity algorithm of integrated feature, and not only used partial form to establish effective information, or some digital error occurred, all can achieved retrieval with database and found the location of candidate forms.

[1]Zanibbi, R., Blostein, D., Cordy, J.R., “A survey of table recognition: Models, observations, transformations, and inferences,” International Journal on Document Analysis and Recognition, Vol. 7, No. 1, pp. 1-16 (2004).
[2]Neves, L.A.P., De Carvalho, J.M., Facon, J., Bortolozzi, F., “Table-form extraction with artefact removal,” Journal of Universal Computer Science, Vol. 14, No. 2, pp. 252-265 (2008).
[3]Nielson, H., Barrett, W., “Consensus-based table form recognition of low-quality historical documents,” International Journal on Document Analysis and Recognition, Vol. 8, No. 2-3, pp. 183-200 (2006).
[4]Zheng, Y., Huiping, L., Doermann, D., “A Parallel-line detection algorithm based on HMM decoding,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, No. 5, pp. 777-792 (2005).
[5]Xi, D., Lee, S.W., “Extraction of reference lines and items from form document images with complicated background,” Pattern Recognition, Vol. 38, No. 2, pp. 289-305 (2005).
[6]Amano, A., Asada, N., “Graph grammar based analysis system of complex table form document,” Proc. Seventh Int'l Conf. on Document Analysis and Recognition, ICDAR, Vol. 2003, pp. 916-920 (2003).
[7]Luo, K., Latifi, S., Taghva, K., Regentova, E., “Recognition and identification of form document layouts,” International Conference on Information Technology: Coding Computing, Vol. 2, pp. 352-356 (2004).
[8]Mandal, S., Chowdhury, S.P., Das, A.K., Chanda, B., “A hierarchical method for automated identification and segmentation of forms,” Proceedings of the International Conference on Document Analysis and Recognition , ICDAR, Vol. 2005, pp. 705-709 (2005).
[9]Liu, J., Jain, A.K., “Image-based form document retrieval,” Pattern Recognition, Vol. 33, No. 3, pp. 503-513 (2000).
[10]He, G.H., Xie, Z.M., Chen, R., “Automatic classification of form features based on neural networks and fourier transform,” Proceedings of the 7th International Conference on Machine Learning and Cybernetics, ICMLC, Vol. 2, pp. 1162-1166 (2008).
[11]Lin, J.Y., Lee, C.W., Chen, Z., “Identification of business forms using relationships between adjacent frames,” Machine Vision and Application, Vol. 9, pp. 56-64 (1996).
[12]Tan, C.L., Huang, W., Yu, Z., Xu, Y., “Imaged document text retrieval without OCR,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 6, pp. 838-844 (2002).
[13]Tzacheva, A., El-Sonbaty, Y., El-Kwae, E.A., “Document image matching using a maximal grid approach,” Proceedings of SPIE - The International Society for Optical Engineering, Vol. 4670, pp. 121-128 (2002).
[14]Duygulu, P., Atalay, V., “A hierarchical representation of form documents for identification and retrieval,” IJDAR, Vol. 5, No. 1, pp. 17-27 (2002).
[15]Marinai, S., Gori, M., Soda, G., “Artificial neural networks for document analysis and recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, No. 1, pp. 23-35 (2005).
[16]Byun, Y., Choi, Y., Kim, G., Lee, Y., “Partial matching: An efficient form classification method,” Proceedings of SPIE - The International Society for Optical Engineering, Vol. 4307, pp. 341-352 (2001).
[17]Chen, L., Feris, R., Turk, M., “Efficient partial shape matching using Smith-Waterman algorithm,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops, No. 4563078 (2008).
[18]鍾國亮，影像處理與電腦視覺，東華書局，台北 (2006)。
[19]Otsu, N., “A threshold selection method from gray-level histogram,” IEEE Trans. Syst., Man, Cybern., Vol. 8, No. 1, pp. 62-66 (1978).
[20]Jain, R., Kasturi, R., Schunck, B.G., Machine Vision, McGRAW-HILL Internation Editions, New York, (1995).
[21]楊淑雅，「表單文件手寫資料欄位擷取之研究」，碩士論文，國立臺灣師範大學，台北 (2006)。
[22]Kuo, S., Cross, G.R., “A two-step string-matching procedure,” Pattern Recognition, Vol. 24, No. 7, pp. 711-716 (1991).
[23]Waterman, M.S. and Eggert, M., “A new algorithm for best subsequence alignments with application to tRNA-rRNA comparisons,” Journal of Molecular Biology, Vol. 197, No. 4, pp. 723-728 (1987).
[24]陳盈全，「改進Smith-Waterman演算法處理序列比對」，碩士論文，中華大學，新竹 (2003)。
[25]王培儀，「利用欄位群聚特徵和四個方向相鄰樹作表格文件分類」，碩士論文，國立中央大學，桃園 (2000)。
[26]施長志，「使用邊角特徵於表格文件辨識之研究」，碩士論文，國立台灣科技大學，台北 (2004)。
[27]Manjunath, B.S., Ohm, J.R., Vasudevan, V.V., Yamada, A., “Color and texture descriptors,” IEEE Transactions on Circuits and Systems for Video Technology, Vol. 11, No. 6, pp. 703-715 (2001).