A Keyword Matching for the Retrieval of Low-Quality Hangul Document Images

다국어 초록 (Multilingual Abstract)

It is a difficult problem to use keyword retrieval for low-quality Korean document images because these include adjacent characters that are connected. In addition, images that are created from various fonts are likely to be distorted during acquisition. In this paper, we propose and test a keyword retrieval system, using a support vector machine (SVM) for the retrieval of low-quality Korean document images. We propose a keyword retrieval method using an SVM to discriminate the similarity between two word images. We demonstrated that the proposed keyword retrieval method is more effective than the accumulated Optical Character Recognition (OCR)-based searching method. Moreover, using the SVM is better than Bayesian decision or artificial neural network for determining the similarity of two images.

참고문헌 (Reference)

1 Lu, Y., "Word searching in document images using word portion matching" 319-328, 2002

2 Jeong, C. B., "Word Extraction from Table Regions in Document Images for Keyword Spotting" 214-223, 2005

3 Doermann, D., "The indexing and retrieval of document images. a survey" 70 (70): 287-298, 1998

4 Rodriguez-Serrano, "Synthesizing queries for handwritten word image retrieval" 45 (45): 3270-3276, 2012

5 Steinwart, Ingo, "Support Vector Machines" Springer- Verlag 2008

6 Strathy, N. W., "Segmentation of handwritten digits using contour features" 577-580, 1993

7 Ohta, M., "Retrieval methods for English-text width misrecognized OCR characters" 2 : 950-955, 1997

8 Gose, E., "Pattern Recognition and Image Analysis" Prentice Hall 1996

9 Kim, Dae Su, "Neural Network Theory and Applications 1" Ha-Tech jeongbo Press 1992

10 Yates, R. B., "Modern Information Retrieval" ACM press 75-82, 1999

11 Jung, M. C., "Machine printed character segmentation method using side profiles" 6 : 863-867, 1999

12 DeCurtins, J., "Keyword spotting via word shape recognition" 270-277, 1995

13 Kim, Soo H., "Keyword Spotting on Korean Document Images by Matching the Keyword Image" 158-166, 2005

14 Park, S. C., "Keyword Spotting on HangulDocument Images Using Two-level Image-to-Image Matching" 79-81, 2005

15 Lu, Y., "Information Retrieval in Document Image Databases" 16 (16): 1398-1410, 2004

16 Mitra, M., "Information Retrieval from Documents" 2 : 141-163, 2000

17 Kim, H. G., "Image-based retrieval of printed Korean words using wavelets" 28 (28): 91-103, 2001

18 Tan, C. L., "Image document text retrieval without OCR" 24 (24): 838-844, 2002

19 Fausett, L., "Fundamentals of Neural Networks" Prentice Hall 1994

20 Marukawa, K., "Document retrieval tolerating character recognition errors-evaluation and application" 30 : 1361-1371, 1997

21 Park, S.C, "Character Segmentation and Keyword Matching for the Retrieval of Low Quality Korean Document Images" Chonnam National University 2006

22 Lazebnik, S., "Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories" 2169-2178, 2006

23 Salton, G., "Automatic Analysis, Theme Generation, and Summarization of Machine-Readable Text" 264 : 1421-1426, 1994

24 Liang, Y., "A synthesised word approach to word retrieval in handwritten documents" 4225-4236, 2012

25 Chen, F. R., "A comparison of discrete and continuous hidden Markov models for phrase spotting in text images" 1 : 398-402, 1995

26 Kwag, H. K, "A Study on Word Segmentation and Attribute Extraction from Document Images" Chonnam National University 2001