• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >A structural skeleton based shape indexing approach for vector images.
【24h】

A structural skeleton based shape indexing approach for vector images.

机译：一种基于结构骨架的矢量图像形状索引方法。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

It is widely believed that the humans recognize natural objects by their shapes. Structural based skeleton based shape indexing (SSBSI) is the process of image retrieval of vector images from databases based on their shape. Given a vector image, it is desirable to retrieve relevant vector images quickly and precisely from a large database. The main goal of this paper is to present a vector image retrieval indexing system that is primarily based on shapes. We propose a new solution, called VIndex, which utilizes shape features and skeletal quantization in 2D vector space. The major contribution of this paper is the discovery of an efficient approach to indexing any given vector image based on skeletal and region geometric features. The proposed solution consists of three parts: a Metafile Compositing Algorithm that removes redundancy, retrieves the command records and translates them into a set of closed contours that delineate contiguous non-overlapping regions; Skeletonization Algorithms that generate geometry skeletal curves; a VIndex Algorithm that subtracts sub-regional and skeletal features and encodes them into a set of data that can be stored into any database system. Experimental results indicate that the proposed indices significantly improve the retrieval efficiency, accuracy and effectiveness.
机译：人们普遍认为，人类通过其形状识别自然物体。基于结构的基于骨架的形状索引（SSBSI）是基于数据库形状对矢量图像进行图像检索的过程。给定矢量图像，希望从大型数据库中快速而准确地检索相关的矢量图像。本文的主要目的是提出一种主要基于形状的矢量图像检索索引系统。我们提出了一种名为VIndex的新解决方案，该解决方案利用了2D向量空间中的形状特征和骨骼量化。本文的主要贡献是发现了一种基于骨架和区域几何特征对任何给定矢量图像进行索引的有效方法。所提出的解决方案包括三部分：图元文件合成算法，该算法去除冗余，检索命令记录并将它们转换为一组封闭的轮廓，以描绘连续的非重叠区域；生成几何骨架曲线的骨架化算法；一种VIndex算法，它减去子区域和骨骼特征并将它们编码为一组数据，可以存储到任何数据库系统中。实验结果表明，提出的指标显着提高了检索效率，准确性和有效性。

著录项

  • 作者

    Song, Mingkui.;

  • 作者单位

    State University of New York at Binghamton.;

  • 授予单位 State University of New York at Binghamton.;
  • 学科 Computer Science.
  • 学位 Ph.D.
  • 年度 2009
  • 页码 159 p.
  • 总页数 159
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 自动化技术、计算机技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号