Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/19318
標題: 基於空間關係之語意影像搜尋系統
Semantic Image Search Based on Spatial Relationships
作者: 孟憲君
Meng, Shian-Jiun
關鍵字: Semantic web
語意網
Image search
OWL
WordNet
Ontology
影像搜尋
OWL
詞彙網路
本體論
出版社: 資訊科學系所
引用: [1] Alexander Maedche and Steffen Staab, “Ontology Learning for the Semantic Web”, IEEE INTELLIGENT SYSTEMS, 2001. [2] E.L. van den Broek, L.G. Vuurpijl, P. Kisters, and J.C.M. von Schmid ,“Content-Based Image Retrieval: Color-selection exploited”, Nijmegen Institute for Cognition and Information, 2002. [3] A.Th.(Guus) Schreiber, Barbara Dubbeldam, Jan Wielemaker, and Bob ielinga, “Ontology-Based Photo Annotation”, IEEE INTELLIGENT SYSTEMS, 2001. [4] Director of the World Wide Web Consortium, http://www.w3.org/People/Berners-Lee/ [5] Michael C. Daconta, Leo J. Obrst and Kevin T. Smith, “The Semantic Web: A Guide to the Future of XML, Web Services, and Knowledge Management”, pp.9-15,ISBN:0471432571. [6] Graham Klyne and Jeremy Carroll, “Resource Description Framework (RDF) Concepts and Abstract Syntax”, W3C Proposed Recommendation, http://www.w3.org/TR/rdf-concepts/, January 23, 2003. [7] Deborah McGuinness and Frank van Harmelen, “OWL Web Ontology Language Overview”, W3C Proposed Recommendation, http://www.w3.org/TR/owl-features/ , August 18, 2003. [8] Jena - A Semantic Web Framework for Java, http://jena.sourceforge.net/ [9] Irena Valova, Boris Rachev, “Retrieval by Color Features in Image Databases”, Budapest, Hungary, Septemper, 2004. [10] Picture Finder Online-Demo, http://134.102.206.80:8080/demo/frameset.jsp/ [11] A.Th.(Guus) Schreiber, Barbara Dubbeldam, Jan Wielemaker, and Bob ielinga, “Ontology-Based Photo Annotation”, IEEE INTELLIGENT SYSTEMS, 2001. [12] A.M. Tam and C.H.C. Leung, “Structured Natural- Language Description for Semantic Content Retrieval”, J. American Soc. Information Science, Sept. 2001. [13] Laura Hollink, Guus Schreiber, Jan Wielemaker, “Semantic Annotation of Image Collections”, Proceedings of the K-CAP 2003 Workshop on Knowledge Markup, 2003. [14] Eero Hyvonen, Hannu Erki, “Ontology-Based Image Annotation and Retrieval”, University of Helsinki Dept. of Computer Science Helsinki 27th April 2005. [15] A lexical database for the English language, http://wordnet.princeton.edu/ [16] Lei Zhang, Fuzong Lin, Bo Zhang,“SUPPORT VECTOR MACHINE LEARNING FOR IMAGE RETRIEVAL”, Department of Computer Science and Technology, Tsinghua University, Beijing, 2001. [17] Zaher AGHBARI, Akifumi MAKINOUCHI, “Semantic Approach to Image Database Classification and Retrieval”, E.E., Kyushu University, 2003. [18] Y.Deng, B.S.Manjunath and H.Shin , "Color Image Segmentation", Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR ''99, Fort Collins, CO, vol.2, pp.446-51, June 1999. [19] JSEG algorithm, http://vision.ece.ucsb.edu/segmentation/jseg/software/ [20] DAMIAN CONWAY, “An Experimental Comparison of Three Natural Language Colour Naming Models”, Proc. East- West International Conference on Human-Computer Interaction, St. Petersburg, Russia, pp. 328-39, 1992. [21] WordNet API, http://sourceforge.net/projects/jwordnet/
摘要: 在現今這個資訊爆發的時代,多媒體影像的資料量與日俱增,如何在充滿各式各樣的圖形中找出使用者真正想要的實在是目前刻不容緩的課題。由於以往以內容為基礎的影像搜尋技術缺乏語意上的支援,常造成搜尋結果並非如使用者所預期,本篇論文提出了在影像中加入語意的支援,使系統對影像能具有更深入的理解,進而提高對影像搜尋的正確性。 本論文基於本體論為基礎,發展出一套具有半自動化標記影像的系統實作。首先,我們利用相似顏色為同一群的分類方法來對影像加以切割。切割完後的每一個區域,可以擷取其顏色的特徵並把這些特徵加入到我們的本體論中。接著,我們就可以只針對我們喜歡的物件加以標記,然後便可以利用我們建立好的本體知識庫所提供空間上的關連性,以及辭彙網路所提供語義上的同義詞類與上義詞類的特性,進而使我們的系統有效的描述影像各物件中動作的行為與空間上的關係。最後,我們針對三種不同的搜尋方式:以關鍵字為基礎、以空間關係為基礎、以簡易自然語言包含空間與動作關係為基礎的方式,作為效能上的比較報告。
As the volume of multimedia images increase drastically on the web, it is important and becoming difficult to search for a targeted picture efficiently. Commonly used content-based image retrieval is short of semantic support. The retrieved resources may be far away from user''s expectation. In this paper, we propose a new image retrieval system incorporated semantics to make the system better in image repository, consequently, to search for an image more accurate. Based on the construction of spatial ontology, we implement a semi-automatic annotation images system. Firstly, images are segmented according to the variations in color. We extract each segmented region, record its color feature and make annotations in the ontology file. Accordingly, we are able to apply the spatial relationships from the established ontology along with synonym and hypernym characteristics from the WordNet for efficient image retrieved. Finally we compare and report the performance of three different searching mechanisms: keyword-based, spatial relationship-based, and simple natural language and action behavior relationship-based.
URI: http://hdl.handle.net/11455/19318
其他識別: U0005-0307200617243700
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-0307200617243700
Appears in Collections:資訊科學與工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館

Show full item record
 
TAIR Related Article
 
Citations:


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.