透過您的圖書館登入
IP:18.216.251.37
  • 學位論文

根據基因演算法產生之空間關係圖之影片尋取

Similarity Retrieval of Videos Based on the Spatial Relation Diagram Generated by the Genetic Algorithm

指導教授 : 梁恩輝

摘要


在過去影片檢索的相關論文中,使用建置於空間關係圖(Spatial Relation Diagram)上的R-tree作為影片尋取的索引結構。而影片之間的相似度則可以用影片中存在的各個物件對,在影片中所發生過的空間關係進行測量。 而在本研究中,我們修改了空間關係在空間關係圖中的位置,產生出修改後的空間關係圖,被命名為改良關係圖(Modified Relation Diagram),將R-tree建置於新的圖上,則可以為影片產生出更好的索引結構,影片尋取的效率也就可以提升。因此,本研究中我們提出了一個使用基因演算法的方法,來為每個物件對產生出接近最佳的索引結構。

並列摘要


An R-tree for each object pair built on a spatial relation diagram of these two objects has been used for retrieving similar videos. The similarity between videos is measured based on the spatial relations between objects. Non-similar data can be filtered at early stages of the process of query using the R-tree. However, a better R-tree can be obtained by modifying the spatial relation diagram such that the efficiency of the process of query can be improved. Hence, we propose to use the genetic algorithm to build a near-optimal R-tree structure for each object pair for similarity retrieval of videos.

參考文獻


[1]K.Sze, K. Lam, G. Qiu, "A New Key Frame Representation for Video Segment Retrieval", IEEE Transactions on Circuits and Systems for Video Technology, vol. 15,no. 9, pp. 1148 - 1155, 2005.
[2]T. Xiang, S. Cong, “Activity Based Surveillance Video content Modeling”, Pattern Recognition, vol. 41, no. 7, pp. 2309 – 2326, 2008.
[3]P. Yu, “Weighted Similarity Retrieval of Video Database”, ICIC Express Letters, vol. 4, no. 5, pp. 2009-2013, 2005.
[4]C. C. Lo, S. J. Wang, L. W. Huang, “Video Retrieval using Successive Modular Operations on Temporal Similarity”, Computer Standards & Interfaces, vol. 26, no. 4, pp. 317–328, 2004.
[5]A. Hanjalic, “Adaptive Extraction of Highlights from A Sport Video Based on Excitement Modeling”, IEEE Transactions on Multimedia, vol. 7,no. 6, pp. 1114 – 1122, 2005.

延伸閱讀