测绘学报

• •    

曲线形态的结构化表达

翟仁健, 武芳,朱丽,朱强   

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-12-28 发布日期:2019-01-01

Structured Representation of Curve Shape

Zhai Renjian, Wu Fang,Zhu Li   

  • Received:1900-01-01 Revised:1900-01-01 Online:2011-12-28 Published:2019-01-01

摘要: 线要素综合不是单纯几何特征上的数据压缩,需要从地理特征上研究线要素蕴涵的空间结构关系。本文以弯曲作为曲线形态的基本结构单元,分析了曲线形态的空间结构特征,提出了一种曲线形态的结构化方法,即利用约束Delaunay三角网模型,结合格式塔原则,对曲线形态进行结构划分,将曲线弯曲划分为基本弯曲和复合弯曲,并对弯曲特征进行识别,给出了弯曲间空间结构关系的探测方法,并利用面向对象模型对曲线形态的结构化知识进行了表示。实验结果证明,与曲线弯曲深度层次二叉树模型相比,本文提出的曲线形态结构化方法在弯曲划分的视觉感受、曲线形态结构化知识表达的完备性、用于曲线化简的实际效果都明显更优。

Abstract: Linear features take an important part of features on map. Generalization of linear features not only is data compress of geometric characteristics, but also needs researching spatial structure relationship of geographical characteristics, which is connotative among linear features. Spatial structural characteristic of curve is analyzed based on bends as a basic structure unit of curve. A structural method of curve shape is proposed, that is using constraint Delaunay triangulation network model, considering Gestalt principle, partitioning the structure of curve shape, dividing curve bends into basic bends and complex bends, and identifying the bend characteristics, method of detecting and expressing spatial structural relationship among bends is given, finally, bend and its structural relationship are class encapsulated by object oriented method. According to experiments’ results,compared with the model of binary tree representation of curve hierarchical structure in depth, the method is more better in curve bend definition, maturity of spatial structural characteristics representation of curve, and line simplification results using the method.