首页 | 本学科首页   官方微博 | 高级检索  
     检索      

特征间互补性结合使用的3D模型检索技术
引用本文:郑伯川,张征.特征间互补性结合使用的3D模型检索技术[J].重庆邮电大学学报(自然科学版),2006,18(1):119-122.
作者姓名:郑伯川  张征
作者单位:西华师范大学,数学与信息学院,四川,南充,637002
基金项目:湖北省教育厅人文社会科学基金
摘    要:三维(3D)模型检索是多媒体信息检索的热点研究问题之一。针对基于单一形状特征的3D模型检索存在 的不足,提出了结合多种形状特征进行模型检索的方法。把模型的面积分布和纬度方向平均半径2种特征被应用 于模型检索。实验结果表明,由于不同形状特征的互补性,采用多种形状特征可以得到更好的检索准确率。

关 键 词:内容的检索  形状的检索  3D模型检索
文章编号:1004-5694(2006)01-0119-04
收稿时间:2005/5/13 0:00:00
修稿时间:2005年5月13日

3D model retrieval technique using complementarity of features
ZHENG Bo-chuan,ZHANG Zheng.3D model retrieval technique using complementarity of features[J].Journal of Chongqing University of Posts and Telecommunications,2006,18(1):119-122.
Authors:ZHENG Bo-chuan  ZHANG Zheng
Institution:School of Mathematics & Information, China West Normal University, Nanchong 637002, P.R.China
Abstract:Content-based 3D retrieval has attracted more and more researchers' interests due to the increasing of 3D models. A main problem in 3D-retrieval is how to extract discriminative shape features to measure the similarity and dissimilarity between different models. However, any shape feature has its advantage as well as its weakness. Differing with previous work, this paper proposed a 3D retrieval algorithm based on hybrid shape features. Two kinds of shape features named area distribution and latitude orientation average radius are extracted and used for similarity calculation. Experiments show retrieving accuracy based on hybrid shape features has a greater improvement than that on single shape feature.
Keywords:content-based retrieval  shape-based retrieval  3D model retrieval
本文献已被 万方数据 等数据库收录!
点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号