|
|||||
|
|
| 一种自适应多阈值多层次三维模型检索的研究 | |
| 引用本文： | 高波涌,潘翔,张三元.一种自适应多阈值多层次三维模型检索的研究[J].江西师范大学学报(自然科学版),2006,30(2):116-119. |
| 作者姓名： | 高波涌 潘翔 张三元 |
| 作者单位： | 1. 中国计量学院,信息工程学院,浙江,杭州,310018;浙江大学,计算机科学与技术学院,浙江,杭州,310027 2. 浙江工业大学,软件学院,浙江,杭州,310014 3. 浙江大学,计算机科学与技术学院,浙江,杭州,310027 |
| 摘 要： | 针对集成多种形状特征、不同的相似度测量用于三维模型检索,提出自适应双阈值的多层次三维模型检索结构.然后将法向量分布直方图和3D Zernike矩两种特征应用于自适应双阈值的三维检索.对一个标准三维模型测试数据库的检索实验表明,采用双阈值多层次检索结构比单特征的检索准确率要高,并且这种非线性组合与固定平均权重和规则的混合特征相比,在提高检索准确率的同时,也能保证检索效率. |
| 关 键 词： | 基于内容的三维检索 自适应 多阈值 |
| 文章编号： | 1000-5862（2006）02-0116-04 |
| 收稿时间： | 2005-11-06 |
| 修稿时间： | 2005-11-06 |
Research on Multilayer 3D Model Retrieval with Adaptive Multilevel-Thresholding |
|
| GAO Bo-yong,PAN Xiang,ZHANG San-yuan.Research on Multilayer 3D Model Retrieval with Adaptive Multilevel-Thresholding[J].Journal of Jiangxi Normal University (Natural Sciences Edition),2006,30(2):116-119. | |
| Authors: | GAO Bo-yong PAN Xiang ZHANG San-yuan |
| Abstract: | To improve the 3D retrieving performance,different shape features and similarity measure are combined.Differing with previous work,a multilayer 3D model retrieval with adaptive multilevel-thresholding is proposed. Two kinds of features named normal histogram and 3D Zernike moment are used for multilayer 3D model retrieval.Experiments using benchmark show that the system has higher accuracy than that uses a single feature.The result shows also that retriving accuracy and efficiency have a greater improvement than that on sum rule of fixed mean weight. |
| Keywords: | content-based 3D retrival adaptive multilevel-thresholding |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |