基于高精度轨迹数据的要素特征点提取研究  

Research on Feature Points Extraction Based on High-precision Trajectory Data

作　　者：梁丽芳 田时雨 王燕云 吴文坛 LIANG Lifang;TIAN Shiyu;WANG Yanyun;WU Wentan(Hebei Province Natural Resources Archives,Shijiazhuang 050031,China;The Second Institute of Surveying and Mapping of Hebei Province,Shijiazhuang 050031,China)

机构地区：[1]河北省自然资源档案馆,河北石家庄050031 [2]河北省第二测绘院,河北石家庄050031

出　　处：《测绘与空间地理信息》2023年第8期36-39,43,共5页Geomatics & Spatial Information Technology

基　　金：国家重点研发计划(2016YFB0502104);河北省技术创新引导计划项目科技冬奥专项(21477603D)资助。

摘　　要：连续运行参考站(CORS)向社会提供高精度位置服务的同时,产生了大量的轨迹数据,这些数据蕴含丰富的地形、地物特征。本文以河北CORS采集的轨迹数据为基础,应用分布式运算的思想,将全省不同测区、不同项目的数据整合、分析,开发了要素特征点提取工具,实现了地理信息要素特征点的自动提取。针对原始观测数据以及地理信息要素空间形态多样、变化不规律的特点,采用轨迹检索、拐点提取和空间密度聚类算法相结合的技术,达到了获取要素细节特征点的目的。测试结果表明,要素形态变化处的关键特征点提取准确率在94%以上,在地形图测绘、控制测量等生产项目中具有重复利用的价值。While providing high-precision location services to the society,the Continuously Operating Reference Stations(CORS) produced a large number of trajectory data.These data contain abundant terrain and object features.Based on the trajectory data collected by Hebei CORS,this paper integrated and analysed the data of different surveying areas and projects in the province.A feature point extraction tool was developed to realize the automatic extraction of geographic information feature points.In view of the various spatial forms and irregular change of the original surveying data,the combination of trajectory retrieval,inflection point extraction and spatial density clustering algorithm is used to obtain the elements and details.The test results showed that the extraction accuracy of key feature points at the change of spatial element form was more than 94%.These feature points will have the value of reuse in topographic mapping,control surveying and other production projects.

关 键 词：要素特征点 CORS 分布式运算 轨迹 空间密度聚类 

分 类 号：P231[天文地球—摄影测量与遥感]