English  |  正體中文  |  简体中文  |  Items with full text/Total items : 53694/88316 (61%)
Visitors : 10176636      Online Users : 32
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/37760

    Title: The extraction of text/graphs from degraded documents
    Authors: Yen, Shwu-huey;Chen, Yi-jin;Lin, Hwei-jen;Wang, Chia-zen
    Contributors: 淡江大學資訊工程學系
    Date: 2004-01-05
    Issue Date: 2010-01-11 14:06:18 (UTC+8)
    Abstract: This paper presents a method for improving the quality of degraded documents by noise removal and text enhancing. Histogram of a degraded document is analyzed to find out the approximate ranges of gray-value for text-, graph-, (i.e. photographs), and background-pixels. After the graph-pixels are identified, they are replaced by the background pixels. Agent-growing method described by S. H. Yen and M. C. Shih (2000) is then applied to smooth the noisy background and a document with clear readable condition for text and background is obtained. At last, graph pixels are recovered to get the final result such that the degraded document now has the text in much better quality and photographs preserved if there is any. Experiments to verify the efficacy of the proposed method and comparison to some existing techniques are also presented.
    Relation: Proceedings of The 10th International Multi-Media Modeling Conference, pp.181-186
    DOI: 10.1109/MULMM.2004.1264984
    Appears in Collections:[Graduate Institute & Department of Computer Science and Information Engineering] Proceeding

    Files in This Item:

    File Description SizeFormat
    The extraction of characters on dated color postcards.pdf313KbAdobe PDF164View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback