透過您的圖書館登入
IP:3.145.156.46
  • 學位論文

根據物件空間關係變化之影片相似尋取-考慮隱含關係

Similarity retrieval of videos based on the movement of the object - considering implied relationship

指導教授 : 梁恩輝

摘要


近年來影片資料量隨著資訊科技進步設備的普及越發增長快速,而進行影片尋取的方法有許多種,使用基於內容的相似尋取是個較適宜的方法之ㄧ。 在影片中,相鄰影格間內因物件移動而產生的空間關係變化是尋取的一個重要資訊,因此本文以影片中兩兩物件的空間關係與其變化為基礎並考慮「隱含關係」來進行影片間是否相似之評量,達到影片相似尋取之目的。最後,我們以範例說明利用我們提出的方法可以根據影格之間物件的空間關係變化作為查詢條件之影片相似尋取。

並列摘要


Recently, the quantity of video information has increased at a high speed due to the improvement of the information technology. There are many methods for video retrieval. Similarity retrieval based on the contents is one of the suitable methods. In the video, the change of the spatial relationship between objects due to the movement of the object provides important information. In this paper, the spatial relationship between objects and its change are used as a base for determining if two videos are similar. In the mean time, the “implied relationship” is also considered in this paper for similarity retrieval of video. Finally, an example is given to demonstrate the process of the proposed method.

參考文獻


[2] S. Chang, Q. Shi and C. Yan, "Iconic indexing by 2-D strings," Pattern Analysis and Machine Intelligence, No. 3, pp.413-428, 1987.
[3] S. Deb and Yanchun Zhang, "An overview of content-based image retrieval techniques," Advanced Information Networking and Applications 18th International Conference, pp.59-64, 2004.
[4] A. Del Bimbo and P. Pala, "Visual image retrieval by elastic matching of user sketches," Pattern Analysis and Machine Intelligence, Vol. 19, No. 2, pp.121-132, 1997.
[6] V. N. Gudivada and V. V. Raghavan, "Content based image retrieval systems," Computer, Vol. 28, No. 9, pp.18-22, 1995.
[7] F. Hsu, S. Lee and B. Lin, "Video data indexing by 2D C-trees," Journal of Visual Languages & Computing, Vol. 9, No. 4, pp.375-397, 1998.

延伸閱讀