测绘学报

• 学术论文 • 上一篇    下一篇

大数据量遥感分类图像的动态矢量化算法

荣岳成   

  1. 北京华迪宏图信息技术有限公司
  • 收稿日期:2011-05-27 修回日期:2011-10-18 出版日期:2012-12-25 发布日期:2013-04-17
  • 通讯作者: 荣岳成

A Dynamic Algorithm of Vectorization for Large Claasified Remote Sensing Image

  • Received:2011-05-27 Revised:2011-10-18 Online:2012-12-25 Published:2013-04-17

摘要:

针对大数据量遥感图像矢量化时间效率低和空间瓶颈问题,采用顶点提取过程中同时进行图斑矢量化的策略,提出一种适应大数据量遥感图像的动态矢量化算法。首先,统计图像中各图斑的顶点个数,将其作为图斑顶点集合能否构成封闭多边形的判定条件;然后,在顶点提取的过程中发现有能够成封闭多边形的图斑顶点集合,立即将其矢量化并释放其所占内存,图斑矢量化时直接将顶点构建成多边形无需生成中间弧段,并即时形成拓扑关系。实验表明,文中算法能快速高效地完成大数据量遥感图像矢量化,速度是ARCGIS(9.2版本)矢量化算法的3~4倍。

关键词: 矢量化, 顶点, 动态化, 拓扑关系

Abstract:

To solve the problem of low efficiency and space bottlenecks of vectorization for large remote sensing image, this paper presents a dynamic algorithm of vectorization for large remote sensing image, based on the policy of verctorizing polygons in the process of vertices extraction. First, the number of vertices in every polygon is counted and it’s used as the factor to judge whether a set of vertices constitute a closed polygon or not; and then during the extraction of vertices, if a set of vertices can constitute a closed polygon, it will be vectorized and its memory will be released. A set of vertices constructs directly into a closed polygon without generating temporary data, created topological structure of vector data during the conversion. Experimental results show that the dynamic algorithm has high efficiency in time and space, and its speed is 3 to 4 times that of ARCGIS (9.2 version).

Key words: vectorization, vertex, dynamism, topological relation