一种边缘分类约束的道路倾斜实景简化方法

A Simplified method of road oblique real scene data with edge classification constraint

针对长距离道路倾斜实景三维模型数据量较大且冗余,传统二次误差度量(QEM)边塌陷法在数据简化时易出现不稳定三角网的问题,该文提出一种边缘分类约束的道路网格数据简化方法。该方法充分结合边的几何结构和属性信息,首先依据塌陷前后法向量的角度变化和相邻边缘的角度特征将边缘边划分为特征边缘边和非特征边缘边,以此计算塌陷点位置,选择最优塌陷边,增大简化率,优化简化效果;然后,基于位置和属性信息划分单一属性边与多重属性边,重新计算多重属性边的塌陷顺序和塌陷点位置,确保简化过程中多重属性边空间位置的一致性。实验结果表明,该文算法较好地保护了三角面的几何特征,得到的边缘数据完整,较好平衡了简化率和简化精度,且效果均优于传统QEM算法。

Aiming at the problem of excessive and redundant data volume in long-distance road oblique 3D models,the traditional quadric error metric(QEM)edge collapse method is prone to unstable triangulation during data simplification,a road mesh data simplification method with edge classification constraints was proposed in this paper.This method fully combines the geometric structure and attribute information of edges.Firstly,according to the angle changes of the normal vector before and after collapse and the angle characteristics of adjacent edges,the border edges were divided into feature border edges and non-feature edge edges.Based on this,the collapse point position was calculated,the optimal collapse edge was selected,the simplification rate was increased,and the simplification effect was optimized;Then,based on position and attribute information,single attribute edges and multiple attribute edges were divided,and the collapse order and collapse point position of the multiple attribute edges were recalculated to ensure consistency in the spatial position of the multiple attribute edges during the simplification process.Experimental results showed that the proposed algorithm effectively protected the geometric features of the triangular surface,obtained complete edge data,balanced simplification rate and accuracy,and performed better than traditional QEM algorithms.