English  |  正體中文  |  简体中文  |  Items with full text/Total items : 51258/86283 (59%)
Visitors : 8010811      Online Users : 100
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/987654321/34157


    Title: 利用時空資訊之影片相似尋取
    Other Titles: Similarity retrieval of videos based on spatio-temporal information
    Authors: 葉智昇;Yeah, Chih-sheng
    Contributors: 淡江大學資訊管理學系碩士班
    梁恩輝;Liang, En-hui
    Keywords: 空間關係;關係序列;最大完全子圖;spatial relation;relation number sequence;maximal complete subgraph
    Date: 2006
    Issue Date: 2010-01-11 04:59:00 (UTC+8)
    Abstract: 隨著多媒體資料庫技術的發展,多媒體資料的數量愈來愈多,其中包括影像資料、影片資料等等。在影片資料庫中,影片中的每一個畫面可視為是一張靜態影像,因此影片的相似尋取技術可由影像相似尋取技術延伸而來。影像內容大致可分為低階視覺特徵(low-level visual feature)和高階關係特徵(high-level relationship feature)。而高階關係特徵中的空間關係(spatial relationships)則是常用衡量兩張影像之間的相似度的標準之一。在過去的研究中,多是以字串表示的方式來描述影像中物件對之間的空間關係,並以這種字串做為相似尋取之根據。在本篇論文中,我們對空間關係做編碼,並提出一個以編碼後的結果表示影片中兩兩物件一連串的空間關係變化的表示法,稱為關係序列(Relation Number Sequence,RNS),並利用關係序列的內容做為相似尋取的依據。此外為了更客觀地表示相似程度,本篇論文以從圖(graph)中找出最大完全子圖(maximal complete subgraph)的方式,提供相似度的值。
    According to the improvement of multimedia techniques there are more and more multimedia data such as image and video. There are also many multimedia database including image database systems and video database systems. In video data, each frame of the video is exactly a still image. The techniques of video data similarity retrieval can be extended from that in image data. Image content can be categorized into low-level visual features and high-level relationship features. In high-level relationship, spatial relationships are usually used in determining the similarity between two images. In the previous researches, spatial relations between two objects are recorded by using string. These strings are then used to determine the similarity between two images.
    In this paper, we encode spatial relations by giving each of them a number. We use these numbers to represent the sequential spatial relation changes between two objects in the video. The sequences of number changes are called Relation Number Sequence (RNS for short). The RNSs are the criteria for determinant the similarity between two videos. In order to provide an objective similarity degree, we use the nodes of a maximal complete subgraph to represent the similarity between two videos.
    Appears in Collections:[資訊管理學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat
    0KbUnknown235View/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