標題: 數位視訊中移動物體切割自動化之研究
A study on automatic moving object segmentation for digital video
作者: 張銘峰
Ming-Feng Chang
陳玲慧
Ling-Hwei Chen
資訊科學與工程研究所
關鍵字: 移動物體切割自動化;物體切割;物體擷取;移動向量;以內容為基礎;視訊壓縮;object segmentation;object retrieval;content-based video compression;content-based;optical flow
公開日期: 2000
摘要: 隨著多媒體應用及以內容為基礎人機互動的普及化,編碼系統的彈性和效率就變得越來越重要。ISO MPEG-4標準吸引許多研究者為以內容為基礎的存取及處理提供技術的解決方案。要滿足這個標準,從視訊中取出移動物體的技術是必須的。本論文將提出一種移動物體切割自動化的方法。這個方法是以移動向量、顏色差值及邊緣點為基礎。文中並提供一些實驗結果以說明此方法的效率優於既存的方法。
With the increasing popularity of multimedia applications and content-based interactivity, the flexibility and efficiency of the used coding systems becomes more and more important. The ISO MPEG-4 standard has attracted much attention recently for providing a technical standardized solution for content-based access and manipulation for these applications. To meet the standard, extracting moving objects from a video sequence is essential. In this thesis, a new automatic video sequence segmentation algorithm that will extract moving objects is presented. The algorithm is based on motion vectors, color difference and edge points. To show the effectives of the proposed method, some experimental results are also given to show that the proposed method is better than an existing method.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT890394057
http://hdl.handle.net/11536/66960
Appears in Collections:Thesis