...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Pattern Analysis and Applications >Geometric algorithm for dominant point extraction from shape contour
【24h】

Geometric algorithm for dominant point extraction from shape contour

机译：从形状轮廓提取优势点的几何算法

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this paper we propose a new algorithm for extracting dominant points from the real contour of a digital shape. A polygonal approximation of the shape can be obtained by the set of dominant points. In the proposed algorithm, in the first step before searching for dominant points, the real contour is made sparse using a geometric concept, named convex deficiency tree. This helps to select a set of candidate points from real contour. In comparison with break points (which are initial points in many algorithms), the set of candidate points is more heuristic and the ratio of them to the all points of the contour is lower. In the second step of the proposed algorithm, the less informative candidate points are removed in an iterative manner. After removing one candidate point, its adjacent positions are searched to find more stable position for its neighbors. The comparative result of the proposed algorithm with others shows its efficiency. The algorithm finds an effective polygonal approximation for digital shapes especially for the real contours, which makes the method more practical.
机译：在本文中，我们提出了一种从数字形状的真实轮廓中提取优势点的新算法。可以通过一组优势点获得形状的多边形近似值。在提出的算法中，在搜索优势点之前的第一步中，使用称为凸缺陷树的几何概念使实际轮廓稀疏。这有助于从真实轮廓中选择一组候选点。与断点（在许多算法中是初始点）相比，候选点集更具启发性，并且它们与轮廓的所有点的比率较低。在提出的算法的第二步中，以迭代方式删除信息量较少的候选点。删除一个候选点后，将搜索其相邻位置以找到其相邻点的更稳定位置。该算法与其他算法的比较结果表明了该算法的有效性。该算法找到了一种有效的数字近似多边形近似，特别是对于真实轮廓，这使得该方法更加实用。

著录项

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