臺灣博碩士論文加值系統

網際網路的快速發展，使得文件的內容不再只局限於文字，還有其它的多媒體資訊，因此如何有效的在網際網路上檢索所需要的多媒體資訊，是一個重要的研究課題。
本論文提出一物件導向影像檢索系統，切割影像中的物件後，使用仔細設計的系統調色盤標示物件，再利用本論文所提出的改良式二維字串(2D-String)方法，記錄物件空間關係，用以解決一般2D String無法解決的旋轉問題，並據以索引及檢索影像。本系統所提出的方法為一兩階段式的影像檢索方法，查詢影像必須與資料庫內的影像具有部份或全部的物件空間關係，再使用低階特徵(空間細部差異及顏色)計算查詢影像與資料庫內的影像的相似度。避免直接使用低階特徵，本系統所提出的方法，可以改善一般內容導向影像資料速度慢的缺點。
除此以外，我們提出一個以XML為基礎的影像檢索系統，透過XML技術描述二元化影像資料內容，進而達到以內容為導向的影像資料檢索系統，證明網站上的影像資料可以透過XML，達到影像資訊檢索與交換的目的。
本研究的主要貢獻包括：(1)讓網站資訊的管理不再只侷限於文字，還包括了影像，更可進一步的利用XML進行其它多媒體資料的管理。(2)以XML制定影像格式，使得異質的影像資料庫系統能夠透過XML進行影像資料的交換，達到整合的目的。(3)提出了一個修正後的2D String的比對模型，加入物件空間關係、旋轉不變的特性。

As the advances of the Internet technologies, more and more books and multimedia documents have been digitized and shared over the Internet. To design an effective and efficient information retrieval scheme, and hence, becomes an important issue for the purpose of querying multimedia data through the Internet.
In this thesis, we propose a new object-based image retrieval system. In our system, the objects of an image are segmented and labeled by the nearest matched color of a carefully designed system color palette. Objects of an image consist of a spatial composition structure and recorded by a modified two-dimensional string structure, which is proposed to solve the rotation-variant problem in the general two-dimensional string methods. The spatial relationships among objects of image are used to index and retrieve images in the database.
Besides, the proposed system is described and managed in XML. The XML technique has been proved to have powerful description capability for interchanging text information through the Internet. We use it to achieve the goal of interchanging multimedia data.

中文摘要 ------------------------------------------------------i
英文摘要 -----------------------------------------------------ii
誌謝 --------------------------------------------------------iii
目錄 ---------------------------------------------------------iv
圖目錄 --------------------------------------------------------v
壹、緒論-------------------------------------------------------1
一、背景說明------------------------------------------------1
二、研究動機------------------------------------------------4
三、研究目的------------------------------------------------5
四、論文架構------------------------------------------------5
貳、文獻探討---------------------------------------------------7
一、內容導向影像資料庫---------------------------------------7
二、符號影像(Symbolic Image)處理----------------------------10
三、XML(Extensible Markup Language)技術--------------------16
參、結合物件空間關係與特徵的影像檢索方法設計---------------------19
一、物件導向影像檢索結構-----------------------------------19
二、改良式的2D-String結構----------------------------------27
三、誤差原因分析-------------------------------------------32
肆、以XML為基礎的影像檢索系統之設計-----------------------------39
一、系統結構----------------------------------------------39
二、XML標籤架構-------------------------------------------43
三、XML文件檢索-------------------------------------------49
四、系統建構步驟-------------------------------------------54
五、異質性影像資料庫整合設計--------------------------------55
伍、實驗步驟與實驗結果的討論------------------------------------57
一、實驗環境----------------------------------------------57
二、實驗步驟----------------------------------------------58
三、實驗結果與分析-----------------------------------------64
陸、結論與未來發展---------------------------------------------68
參考文獻 -----------------------------------------------------70

[1] S.Y Lee, M.K Shan,1990, "Access method of image database,Internet", Pattern Recognition and Artificial Intelligence 4, pp. 27-44.
[2] Bilge Gunsel, A. Murat Tekalp,1996, "Similarity Analysis for Shape Retrieval by Example", IEEE, Processings of ICPR''96, pp300-304.
[3] R.Mehrotra and J.E.Gray,1993, "Feature-Based Retrival of Similar Shapes," 9th International Conferences on Data Enginnering,, pp.108-115.
[4] Seung-Hoon Lee, Gi-Hwa Jang, Su-Hyun Lee, Sung-Hwan Jung, Yong-Tae Woo, 1997, "A Content-Based Image Retriveal System Using Extend SQL in RDBMS" , Communications and Signal Processing, International Conference on Information , pp.9-12, Sep.
[5] Flickner, M., Sawhnery, H., Niblack, J., Lee, D. Petkovic, D., Steele, D, Yanker, P., 1995, "Query by Image and Video Contentl: Yhe QBIC System", IEEE Computer, pp23-32, September.
[6] A. Pentland, R. W. Picard, and S. Sclaroff, 1994, "Photobook: Content-Based Manipulation of Image Database", Proc,SPIE Storage and Retrieval Image and Video Database II,N0. 2185,Feb.
[7] S.K Chang, Q.Y Shi, and C.W Yan, 1984, "Iconic Indexing by 2-D Strings", IEEE Trans. Pattern Anal. Mach. Intell., vol PAMI-6, pp.475-484.
[8] C.C Chang, J.C Lin,1990, "A fast match access schema for symbolic picture", J. Electr. Engineering 33, pp.129-137.
[9] C. Faloutsos,1985, "Signature files: Design and performance compare of some signature extraction metrhods", ACM SIGMOD, pp. 63-82.
[10] R. Sacks-Davis and A. Kent,1987, "Multikey access method based on superimposed coding techniques", Database System 12, , ACM Trans, pp.655-696.
[11] Ye-In Chang, Bi-Yen Yang, 1997, "Efficient access method for image dtabase", Information Processing Letters 64 , pp.95-105.
[12] S.Y Lee and F.J Hsu, 1990, ŖD C-string：A new spatial Knowledge representation for image database systems", Pattern Recognition, 23, pp. 1077-1087.
[13] S.Y Lee, M.K Shan, and W.P Yan, 1989, "Similarity retrieval of iconic image database", Pattern Recognit, 22, 6, pp675-682, Nov.
[14] W.I. Grosky and Y. lu, 1986, "Iconic Indexing Using Generalized Pattern Matching Techniques", Computer Vision Graphics Image Process., Vol. 35, pp. 383-403.
[15] G. Bordogna, I. Gagliard, D. Merelli, P. Mussio, M. Padula, M. Proui, 1989, "Iconic Queries on Pictorial Data", IEEE Workshop on Visual Languages, pp. 38-42, Oct.
[16] S.k Chang, C.W. Yan, D.C. Dimitroff, and T.Arnd , 1988, "An Intelligent Image Database System", IEEE Trans. Soft. Eng., vol14,no. 6, pp. 681-688, May.
[17] N. Roussopoulos, and D. Leifker, 1984, "An Introduction to PSQL:A Pictorial Structured Query Language", Proc. IEEE Workshop Visual Lang, pp. 77-87., Dec.
[18] Venkat N. Gudivada, and Vijay V. Raghavan, 1995, "Design and Evaluation of Algorithms for Image Retrival by Spatial Similarity", Transactions on Information Systems, Vol 13, NO 2, ACM ,pp.114-144,April.
[19] Essam A. El-Kwae, and Mansur R. Kabuka, 1999, "A Robust Framework for Content-Based Retrieval by Spatial Similarity in Image Databases", Transactions on Information Systems, Vol 17, NO. 2, ACM ,pp.174-198,April.
[20] S.K Chang, and E. Jungert, 1991, "Pictorial data management based upon the theory of symbolic projection", Visual Lang., Comput., 2, 3, pp.195-215, Sep.
[21] P.W Huang, and Y.R Jean, 1994,"Using 2D c+-String as spatial knowledge representation for image database systems", Pattern Recognit, 27, 9, pp.1249-1257.
[22] C.C Chang, and S.Y Lee, 1991, "Retrieval of similar pictures on pictorial database", Pattern Recognit. 24, 7, pp.67-681.
[23] T. Bray, J.Paoili, and C.M Sperberg-McQueen, 1998,"Extensible Markup Language(XML) 1.0", World Wide Web Consortium Recommandation, Feb,.
[24] Metthew Fuchs, 1998, "Buliding an Information System from the Web", IEEE Proc. 31st Annual Hawaii International Conference on System Sciences, pp.14-23.
[25] Aaron Weiss, 1999, "XML gets down to business", Mixed Media, ACM, pp36-43, Sep.
[26] Robert J. Glushko, Jay M. Tenenbaum, and Bart Meltzer, 1999, "An XML framework for agent-based E-commerce", Communication, Vol. 42, No .3, ACM, pp.106-114, March.
[27] Steve Widergren, Amold deVod, and JunZhu, , 1999 ,"XML for Data Exchange", pp.840-842,, IEEE, May,.
[28] Andrew V.Royappa, 1998 ,"Implementing Catalog Clearinghouse with XML and XSL", pp616-623, ACM.
[29] J. Clark, S. Deach,1998, "Extensible StyleSheet Language(XSL) Version 1.0", World Wide Web Consortium Working Draft,August.
[30] Shyi-Chyi Cheng and Chen-Kuei Yang,1999, “Fast and Novel Techniques for Color Quantization Using Reduction of Color Space dimensionality”, submitted for publication.
[31] Apparao, V et al., 1998 ,"Document Object Model Specification", W3c Working Draft 16.
[32] William J Pardi ,1999,XML in Action,,鄒正平譯,華彩軟體股份有限公司.
[33] Paul Spencer,1999,Professional XML Design and Implementation,Wrox Press Ltd.