淡江大學機構典藏:Item 987654321/52810
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 55184/89457 (62%)
造訪人次 : 10665117      線上人數 : 22
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/52810


    題名: Fast Image Restoration Method Based on the Multi-Resolution Layer
    作者: Hsieh, Ching-Tang;Chen, Yen-Liang;Hsu, Chih-Hsu
    貢獻者: 淡江大學電機工程學系
    關鍵詞: Wavelet;Image Restoration;Wireless;BNM and WSBNM
    日期: 2009-12-01
    上傳時間: 2010-12-01 10:34:12 (UTC+8)
    出版者: 臺北縣:淡江大學
    摘要: When transmitted through a poor quality network or stored on an unstable storage media, block-based code images will experience the block loss. To restore damaged images suffering from block loss, Best Neighborhood Matching and Jump and Look-Around BNM provide the most effective image restoration. However, while BNM offers good restoration quality, it requires a large calculation time. By “JUMP” method, JLBNM can effectively shorten the computation time but this comes at the cost of a loss in quality. We have therefore proposed a new image inpainting technique that uses theWavelet Domain to deliver fast computation time and high restoration quality Wavelet Stage BNM. Our proposed reconstruction algorithm includes three optimization techniques change of analytical domain, consideration of texture composition and a new decision-making mechanism: DirectionalWaveletWeighted Method. Theoretical analysis and experimental results demonstrate our method delivered fast computation time and high restoration quality.
    關聯: 淡江理工學刊=Tamkang Journal of Science and Engineering 12(4),頁439-448
    DOI: 10.6180/jase.2009.12.4.08
    顯示於類別:[電機工程學系暨研究所] 期刊論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    1560-6686_12-4-8.pdf505KbAdobe PDF144檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

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