English  |  正體中文  |  简体中文  |  Items with full text/Total items : 52343/87441 (60%)
Visitors : 9113613      Online Users : 265
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/987654321/102577

    Title: 累進式全景圖影像修補
    Other Titles: Progressive panoramic image completion
    Authors: 葉昊禹;Yeh, Hao-Yu
    Contributors: 淡江大學資訊工程學系碩士班
    Keywords: 影像修補;馬可夫隨機場;優先權式可靠度傳遞;信賴度值;image completion;Markov Random Field;Priority Belief Propagation;confidence value
    Date: 2014
    Issue Date: 2015-05-04 09:59:17 (UTC+8)
    Abstract: 本論文提出一基於優先權式信賴度傳遞的方法能夠修補缺失區域位在影像四周的情況。要能夠達到這個目的,本方法重新定義馬可夫隨機場中節點的產生方式,並且使用漸進式方法一層一層的去修補影像。同時為了降低其運算量,我們先限制取得候選區塊的範圍之後再使用亮度和飽和度來對馬可夫隨機場中之每個節點的候選區塊做篩選。最後為了產生一張視覺上美觀的結果圖,因此在修補的階段我們加入了節點的目標區塊和候選區塊間之相似度與信賴度值這兩個條件來判斷每個節點是否可以修補。
    In this paper we propose a priority belief propagation based method to complete the missing region located on the periphery of the input image. To reach this goal, we redefine the nodes of Markov Random Field (MRF) and use onion-based approach to complete the image. Meanwhile, avoiding the heavy computation, we restrict the range to getting candidates for MRF. Then, applying intensity and saturation as criteria to filter the candidate of each node of MRF. Finally, we combine the similarity between MRF node’s target and best candidate patch and confidence value to decide whether the node is suitable to complete or not. From the experiments shows that our method successfully produces the visual plausible results.
    Appears in Collections:[資訊工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat

    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