期刊文献+

一种新的基于混合图的图像结构关系描述 被引量:2

A New Image Structures Description Based on Mixed Graph
下载PDF
导出
摘要 针对关系树结构描述图像结构的不完整性,提出一种可用于描述图像区域拓扑关系的混合图结构。利用区域分割方法将图像分割成若干区域,并确定区域间拓扑关系后,建立混合图结构描述这种关系。混合图中的结点代表各区域,有向边表示区域间的包含关系,无向边表示区域间的相邻关系。该结构不仅具有关系树结构表达图像的能力,还考虑了区域间是否相邻,更完整地表示图像各区域的位置关系。可应用于需要精确匹配原图像目标的图像检索、图像识别等。 To aim at the problem of the incompleteness in describing images with relationship tree, a new structure based on mixed graph was presented in this paper to characterize the regional topological relations of images. We employ a region-based method to segment an image and then create the mixed graph structure after obtaining the topological relations of the segmented image. In this structure, nodes represent regions, directed and undirected edges mean inclusion and adjacency relations respectively. The structure not only inherits the merits of the relationship tree, but also effectively describes the adjacency, which is an important relation yet often omitted in image analyzing. This method can be applied in accurate image retrievals or image recognitions.
出处 《福建师大福清分校学报》 2009年第2期47-51,共5页 Journal of Fuqing Branch of Fujian Normal University
基金 福建省科技厅基金项目(2006F5027) 福建省教育厅基金资助项目(JB06102)
关键词 图像描述 区域分割 混合图 关系树 Image description, region segmentation, mixed graph, relationship tree
  • 相关文献

参考文献2

二级参考文献1

同被引文献17

  • 1柳稼航,杨建峰,单新建,尹京苑.一种基于优先搜索方向的边界跟踪算法[J].遥感技术与应用,2004,19(3):209-213. 被引量:26
  • 2严蔚敏,吴伟民.数据结构[M].北京:清华大学出版社,2003.
  • 3Eshera M A, Fu K S. An image understanding system u-sing attribued symbolic repressention and inexact graphmatching[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence, 1986,8(5) :604-618.
  • 4Rocha J, Pavlidis T. A shape analysis model with applica-tions to a character recognition system [ J ]. IEEE Transac-tions on Pattern Analysis and Machine Intelligence, 1994,16(4);393-404.
  • 5Bunke H, Messmer B T. Efficient attributed graph matc-hing and its application to image analysis [ C]// Proceed-ings of the 8 th International Conference on Image Analysisand Processing. 1995:45-55.
  • 6Shapiro L G, Haralick R M. Algorithms for inexact matc-hing[C]// 5th International Conference on Pattern Recog-nition. 1980 :202-207.
  • 7Lin Yang, Teng Zhong Jian. A novel approach for imagerepresentation and matching based on mixed graph structure[C ] //International Conference on Computational Intelli-gence and Software Engineering. 2009: 1-4.
  • 8Hlaoui A,Wang S. A new algorithm for inexact graph matc-hing[ C]// Proceedings of 16th International Conference onPattern Recognition. 2002: 180-183.
  • 9Hlaoui A, Wang S. Image Retrieval Systems Using GraphMatching [ R]. Research Report, No. 275, D6partement deMath6Matiques et d ’ Infoimatique, University de Sher-brooke ,2001.
  • 10Bunke H, Shearer K. A graph distance metric based on themaximal common subgraph[ J]. Pattern Recognition Let-ters, 1998,19(34):255-259.

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部