CASIA OpenIR  > 智能制造技术与系统研究中心  > 多维数据分析
Locating object using hierarchicl hausdorff distance matching
Wei Y.F.; Jiang D.L.; Peng S.L.
2007
Conference Name8th International Conference on Signal Processing ICSP 2006
Conference Date2006/11/16-2006/11/20
Conference Place中国Guilin China
AbstractIn order to locate all the objects of a model in a large image we proposed an effective Hausdorff distance (HD) matching algorithm that is based on coarse-to-fine hierarchical matching. In constructing the hierarchical pyramidal structure the decimation mode of decomposition and the number of hierarchical levels are determined adaptively by the edge model. Thus the proposed algorithm can decrease the number of losing objects at the same time sharply speed up the search when locating objects in an image. For integrate circuit’s (IC) micro images which are mainly composed of vertical and horizontal linear edges the algorithm shows better performance. The experiments also illustrate good matching results with other images.
KeywordHausdorff Distance Hierarchical Structure decimation Mod
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/12886
Collection智能制造技术与系统研究中心_多维数据分析
Corresponding AuthorWei Y.F.
Recommended Citation
GB/T 7714
Wei Y.F.,Jiang D.L.,Peng S.L.. Locating object using hierarchicl hausdorff distance matching[C],2007.
Files in This Item:
File Name/Size DocType Version Access License
Locating object usin(150KB)会议论文 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wei Y.F.]'s Articles
[Jiang D.L.]'s Articles
[Peng S.L.]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wei Y.F.]'s Articles
[Jiang D.L.]'s Articles
[Peng S.L.]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wei Y.F.]'s Articles
[Jiang D.L.]'s Articles
[Peng S.L.]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

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