WORD RETRIEVAL FROM OLD LANNA MACHINEPRINTED DOCUMENT BY MATCHING KEYWORD IMAGE
Keywords:
Lanna document images, word retrieval, keyword image matching, synthetic Lanna word image, Lanna word searchingAbstract
In this paper, we propose a method for word retrieval from old Lanna machine-printed document images based on keyword image matching. The research aims at detecting given keywords given by the user and locates it them on document images. The first step successfully creates a synthetic keyword image. Then, we extract a list of candidate words from the document image by using three 3 extraction methodologies: pixel-based sliding window template matching, block-based sliding window template matching, and word feature matching. Next, a feature vector of each candidate word image is extracted using a window-based feature extraction. Finally, all relevant words are retrieved by comparing the similarity between the keyword and the word image feature vector. The similarity between two 2 feature vectors is evaluated using two 2 methods: sub-window similarity and Euclidean distance. The experimental results show that the method using a combination of word feature matching and Euclidean distance provides the best performance. It is shown that the proposed method is feasible, valid, and effective for Lanna word image searching.
References
Ataer, E., and Duygulu, P. (2006). Retrieval of Ottoman documents. Proceedings of the 8th ACM International Workshop on Multimedia Information Retrieval; October 26 - 27, 2006; Santa Barbara, California, USA, p. 155-162.
Chaijaruwanich, J. (2009). Wannapim Lanna. Chiang Mai University, Chiang Mai, Thailand, p. 5-185.
Doermann, D. (1998). The indexing and retrieval of document images: a survey. Comput. Vis. and Image Understanding., 70(3): 287-298.
Haralick, R. M., and Shapiro, L. G. (1992). Computer and Robot Vision. 1st ed. Addison-Wesley Longman Publishing Co., Inc., Boston, MA, USA, 672p.
Kim, S. H., Park, S. C., Jeong C. B., Kiml, J. S., Park, H. R., and Lee, G. S. (2005). Keyword Sspotting on Korean document images by matching the keyword Iimage. Proceedings of the 8th International Conference on Asian Digital Libraries (ICADL 2005); December 12-15, 2005; Bangkok, Thailand, p. 158-166.
Konidaris, T., Gatos, B., Ntzios, K., Pratikakis, I., Theodoridis, S., and Perantonis S. J. (2007). Keyword-guided word spotting in historical printed documents using synthetic data and user feedback. Int. J. on Doc. Anal. and Recognition (IJDAR)., 9(2): 167-177.
Lu, Y., and Tan, C. L. (2002). Word searching in document images using word portion matching. Proceedings of the 5th International Workshop on Document Analysis Systems; August 19-21, 2002; Princeton, NJ, USA, p. 319-328.
Maneechedta, W. (1994). Enculturation process on Lanna dhama alphabets., Master’s [M.Ed. Thesis]. Department of Nonformal Ecucation, Graduate School, Chiang Mai University, Chiang Mai, Thailand, 79p. (in Thai)
Marukawa, K., Hu, T., Fujisawa, H., and Shima, Y. (1997). Document retrieval tolerating character recognition Eerrors-Eevaluation and application. Pattern Recognition., 30(8): 1361-1371.
Meshesha, M., and Jawahar, C. V. (2008). Matching word images for content-based retrieval from printed document images. Int. J. of Doc. Anal. and Recognition (IJDAR)., 11(1): 29-38.
Otsu, N. (1979). A threshold selection method from gray-level histograms. IEEE T. on Syst., Man and Cybernetics., 9(1): 62-66.
Pansook, A. (2008). Status and role changes of the Lanna language., Master’s [M.A. thesis]. Department of Lanna Language and Literature., Graduate School, Chiang Mai University, Chiang Mai, Thailand, 187p. (in Thai)
Prongthura, N. (1982). Dramma scripts of Northern Thailand., Master’s [M.A. Tthesis]. Department of Oriental Language, Graduate School, Silpakorn University, Bangkok, Thailand, 313p. (in Thai)
Saengbun, P. (2016). Font Lanna "LN TILOK". Available from: http://art-culture.chiangmai.ac.th/fontlanna. Accessed date: Sep 30, 2016.
Senda, S., Minoh, M., and Ikeda, K. (1993). Document image retrieval system using character candidates generated by character recognition process. Proceedings of the 2nd International Conference on Document Analysis and Recognition (ICDAR ’93); October 20-22, 1993; Tsukuba, Japan, p. 541-546.
Yadav, S., and Sawarkar, S. (2009). Retrieval of information in document image databases using partial word image matching technique. Proceedings of the IEEE International Advance Computing Conference (IACC 2009); March 6-7, 2009; Patiala, India, p. 552-557.
