一种基于平行系数的双线主干道识别方法

发布时间：2018-05-08 01:24

本文选题：制图综合 + 城市道路网　；参考：《武汉大学学报(信息科学版)》2017年08期

【摘要】：城市主干道路的识别和提取是路网综合的关键步骤,而双线道路则是大比例尺地图数据中道路综合的难点。针对城市双线主干道识别问题,基于Gestalt视觉准则构建候选双线主干道线对的约束条件,提出了一种基于平行系数的双线主干道识别方法。首先对道路网数据进行拓扑处理,然后借助道路匹配思想,结合Hausdorff(HD)距离匹配方法识别出可能构成双线主干道的候选线对。再对候选线对进行平行系数计算,当平行系数满足阈值条件时,就判定该线对是构成双线主干道的弧段。最后根据构成双线主干道路段间的空间关系,将已识别的弧段连接成整条道路。实验证明,选取典型样本方法正确设置阈值后,该方法能有效地提取道路网中的双线主干道。
[Abstract]:The identification and extraction of urban trunk road is the key step of road network synthesis, while the dual road is the difficulty of road synthesis in large scale map data. Aiming at the problem of urban dual trunk road identification, a method based on parallel coefficient is proposed to identify the dual trunk road, which is based on the Gestalt vision criterion to construct the constraint condition of the candidate dual trunk road pair. Firstly, the road network data is topologically processed, and then, with the aid of the road matching idea and the Hausdorff HDD distance matching method, the candidate line pairs that may form the two-line main road are identified. Then the parallel coefficient of the candidate line pair is calculated. When the parallel coefficient meets the threshold condition, it is determined that the line pair is the arc of the two-line main road. Finally, the identified arcs are connected into the whole road according to the spatial relationship between the sections of the two-lane main road. The experimental results show that this method can effectively extract the two-line main road in road network after selecting the typical sample method to set the threshold correctly.
【作者单位】：信息工程大学地理空间信息学院;苏州工业园区格网信息科技有限公司;中国洛阳电子装备试验中心;西安测绘信息技术总站;
【基金】：国家自然科学基金(41471386,41101362,41301524) 信息工程大学地理空间信息学院学位论文创新与创优基金(XS201509)~~
【分类号】：P208

【参考文献】