基于空間關系的子區(qū)域最佳匹配算法
Optimal Matching Based on Spatial Relationship in Local Regions
-
摘要: 該文提出了一種基于空間關系的圖像檢索算法子區(qū)域最佳匹配法。這種算法首先根據(jù)符號投影模型，將圖像內(nèi)目標抽象為帶有多個屬性的圖標。然后，利用搜索鄰節(jié)點的等角度區(qū)域最近原則、劃分子區(qū)域的中心點法、圖論中的廣度優(yōu)先算法進行圖標匹配。最后，根據(jù)所定義的局部區(qū)域相似距離及圖間距離來計算圖像與圖像之間的相似度，進而完成基于圖像內(nèi)部空間關系的檢索。該文以這種算法為核心，構建了一個籃球站位分析演示系統(tǒng)，并用實際數(shù)據(jù)驗證了所提出的基于空間關系的圖像檢索方法的有效性。Abstract: A new algorithm named optimal matching based on spatial relationship in local regions for content-based image retrieval is proposed in this paper. The algorithm first employs the symbol mapping model to reduce a symbolic image to a l-D representation, and then uses the principle of nearest distance within equal angle region for searching nearest nodes, takes the center point for dividing images into sub-regions, and applies the breadth-first traversal algorithm in graph theory to perform graph matching. Finally, it uses the defined similar distances between local regions and among different graphs to compute the similarity and to retrieve images based on inner spatial relationship. Based on these algorithms, a prototype system for analyzing the position of players in basketball game is constructed. The effectiveness of image retrieval based on spatial relationship is verified in this system by using some real data.
-
章毓晉.基于內(nèi)容的視覺信息檢索.北京:科學出版社,2003,第1章,第3章.[2]Wang Y H. A spatial relationship method supports image indexing and similarity retrieval. In: Multimedia Systems and Content-Based Image Retrieval, Deb S, ed. Hershey, London,Melbourne, Singapore: Idea Group Publishing, 2004, Chapter 12,277 - 301.[3]田福海,等.現(xiàn)代籃球訓練藝術.上海:上海教育出版社,1993,第5章.[4]Sciascio E D, Donini F M, Mongiello M. Spatial layout representation for query-by-sketch content-based image retrieval.Physical Review Letters, 2002, 23(13): 1599 - 1612.[5]http:∥wwwqbic.almaden.ibm.com/.[6]http:∥www.ctr.columbia.edu/VisualSEEK/.[7]章毓晉.基于內(nèi)容的視覺信息檢索.北京:科學出版社,2003,第7章.[8]Del Bimbo A. Visual Information Retrieval. San Francisco:Morgan Kaufmann Publishers, Inc., 1999, Chapter 5.[9]Preiss B R. Data Structures and Algorithms with Object-Oriented Design Patterns in C++. John Wiley Sons, 1998.[10]章毓晉.圖象工程(下冊)--圖像理解與計算機視覺.北京:清華大學出版社,2000,第8章.[11]Albuz E, Kocalar E, Khokhar A A. Scalable color image indexing and retrieval using vector wavelets[J].IEEE Trans. on Knowledge and Data Engineering.2001, 13(5):851- -
計量
- 文章訪問數(shù): 2939
- HTML全文瀏覽量: 103
- PDF下載量: 1054
- 被引次數(shù): 0
下載:
