臺灣博碩士論文加值系統

視訊（Video）是多媒體（Multimedia）中重要的一種媒體。由於視訊本身包含了聲音與影像的部分，較以往傳統的文字或圖片等傳播媒介更為生動與豐富，而且可以更完整、更直接傳達訊息，因此視訊被廣泛的應用在我們的日常生活當中。也正因為視訊的內容豐富，資料量過於龐大，使得要儲存與處理視訊資料需要非常多的記憶空間以及計算能力，由於在視訊資料上並沒有作一些適當的索引機制，所以當使用者想要檢索視訊的內容時，往往必需要將視訊從頭到尾的重新瀏覽，如此一來不但耗費時間與精神，查詢的結果也有可能不盡理想，使得視訊的檢索變成是一件非常沒有效率的工作。因此，一個好的視訊索引（Video Indexing）方法在視訊檢索上扮演了重要的角色。
本論文提出一種利用視訊中各個物件（Object）的空間關係（Spatial Relation）以及時間關係（Temporal Relation），分別為物件命名，然後產生物件表格（Object table）以及影片表格（Frame table），這些表格分別記錄了物件在視訊中出現的時間與位置。透過使用者所設定的查詢條件將物件表格與影片表格中的資料作一篩選，最後找到使用者所欲查詢的資料。本論文所提的視訊索引法可以建立起物件之間的關係，如此一來使用者就可以用口語化的方式來進行視訊的查詢，這樣的查詢不但有效率，而且與使用者之間也較具有親和力。本論文也舉出一些實際的例子，透過與各種索引方式的比較，證明了本論文所提出的方法較其他方法優良。

Video is one of the important medium of multimedia. By containing image and audio, video can transmit the message more plentifully, completely and directly than the conventional media like characters and pictures. Therefore, video is applied in our daily life extensively. With a huge volume of data in video database, it greatly requires a large amount of memory and computing power for storage and processing. Due to no proper searching strategies, users need to browse the video from start to the end again as they want to query the content of video. In this way, searching becomes time-and-mental consuming and inefficient. The result of searching is not so ideal; therefore, a good video indexing plays a significant role in video query.
In this thesis, an approach−by using “Spatial relation” and “Temporal relation” in every object of video is proposed. Firstly, the objects are named and both of the “Object table” and “Frame table” are generated. These two tables record time and places of objects appearing in video. Through the query qualification being set by users, the object table and frame table can be selected. At last, the system can find out the data being searched. The video indexing, which is proposed in this thesis can set up the relation among objects. Consequently, the users can search the video data plainly and efficiently. It’s more intimate between video query and users. Through comparing to all kinds of searching approaches and taking some actual examples, it approves this approach is better than others.

第一章 緒論……………………………………1
1.1 研究動機………………………………………1
1.2研究目的與方法……………………………… 3
1.3 論文架構………………………………………5
第二章 相關研究………………………………6
2.1 時間關係………………………………………8
2.2 空間關係………………………………………15
2.2.1 最小邊界矩形法（Ｍinimum Bounding Rectangles，M.
Ｂ.R.） ………………………… 15
2.2.2 169種空間關係……………………………17
2.2.3二維字串表示法（2D string）…………18
2.3綜合時間與空間的R-tree索引法………………31
2.3.1 Simple spatial and temporal indexing scheme.....31
2.3.2 Unified spatio-temporal indexing schem..........33
2.3.3本節範例……………………………………34
第三章 結合時間與空間關係的視訊索引法.....43
3.1 時間關係與空間關係的定義…………………43
3.1.1 時間關係表示子...................44
3.1.2 空間關係表示子...................44
3.2 物件命名原則........................45
3.3 物件的時空描述表示法.................47
第四章 範例說明.........................54
第五章 結論與未來研究方向................71
5.1 結論................................71
5.2 未來研究方向........................72
參考文獻................................74

[ 1] M. Vazirgiannis.; Y. Theodoridis.; T. Sellis, “Spatio-Temporal Composition in Multimedia”, Multimedia Software Development, 1996. Proceedings., International Workshop on , 1996 , Page(s): 120 —127
[ 2] S. Chang ; Q. Shi ; C. Yan, “Iconic indexing by 2D strings”, IEEE Transactions on Pattern Analysis and Machine Intelligence. 9(3): 413-428,May 1987.
[ 3] Suh-yin Lee and Fang-jung Hsu. “2D C-string: A new spatial knowledge representation for image database systems”, Pattern Recognition,23(10): 1077-1087, 1990
[ 4] J.F. Allen, “Maintaining Knowledge about Temporal Intervals”, Communications of the ACM,vol.26(11),832-843,November 1983.
[ 5] Ma, J.; Knight, B.; Peng, T. “Representing Temporal Relationships Between
Events and Their Effects”, Temporal Representation and Reasoning, 1997. (TIME '97), Proceedings., Fourth International Workshop on , 1997 , Page(s): 148 —152
[ 6] Pissinou, N.; Radev, I.; Makki, E.; Campbell, W.J., “A Topological-Directional Model for the Spatio-Temporal Composition of Video Objects”, Reasearch Issues In Data Engineering, 1998. 'Continuous-Media Databases and Applications'. Proceedings., Eighth International Workshop on , 1998 , Page(s): 17 —24
[ 7] D. Papadias, Y. Theodoridis, “Spatial Relations, Minimum Bounding Rectangles,and Spatial Data Structures”, International Journal of Geographic Information System, 1996.
[ 8] Liang, R.Z.; Venkatesh, S.; Kieronska, D. “Video Indexing by Spatial Representation”Intelligent Information Systems, 1995. ANZIIS-95. Proceedings of the Third Australian and New Zealand Conference on , 1995 , Page(s): 99 -104
[ 9] Ming-Chwen Yang. 2D B-string Representation and Access Methods of Image Database. 交大碩士論文.1990
[10] Brinkhoff ,T. Kriegel, H. P. and Schneider, R.” Comparison of approximations of complex objects used for approximation-based query processing in spatial database systems ”. Proceedings of 9th international conference on data engineering (Computer Society) pp.40-49.
[11] D. Papadias, Y. Theodoridis, T. Sellis, M. Egenhofer,” Topological Relations in the Word of Minimum Bounding Rectangles: a study with R-trees”. Proceedings of ACM SIGMOD Conference, May,1995.
[12] E. Jungert, “Extended symbolic projection used in a knowledge structure for spatial reasoning” 4th BPRA Conference on Pattern Recognition. Springer,Combridge, March 1998.
[13] S.K. Chang, E. Jungert and Y. Li, “Representation and retrieval of symbolic pictures using generalized 2D string” Technical report, University of Pittsburg, 1988.
[14] Sheau-Chuang Wu, Similarity Retrieval of Image Database Based on Decomposed Objects.淡江大學資訊管理研究所碩士論文,1995.
[15] Theoderidis, Y.; Vazirgiannis, M.; Sellis, T.”Spatio-Temporal Indexing for Large Multimedia Applications.”Multimedia Computing and Systems, 1996., Proceedings of the Third IEEE International Conference on , 1996 , Page(s): 441 -448