摘要
针对现有管理方法无法有效支持车载移动激光扫描多源数据联合后处理的问题,提出了一种针对城市街区尺度下海量车载多源数据的管理与调度方法。在分析各类数据空间分布特性基础上,通过有机联合瓦片索引与随机采样八叉树等空间索引结构,并结合数据库进行元数据管理及动态调度控制,建立一套面向车载移动扫描点云、全景影像、POS轨迹、DOM等海量异构多源数据的高效管理及调度机制。实验证明,该方法能够支持城市街区尺度下海量车载多源数据的高效调度、交互查询、动态更新等过程。
Aiming at the absence of efficient management method for the combined processing of vehicle-borne mobile scanning multi-source data,a novel organization and swapping approach is presented, aiming at the urban-block scale massive data. In this paper, based on the spatial distribution analysis of concerning data, a novel mechanism is established to manage the massive heterogeneous data effectively, including the mobile scanning point cloud, panoramic image, trajectory and DOM, etc. In this method, combined with the relational database for meta data management, different indexing structures including the random sampling octree are integrated appropriately to support the well performance of data dispatching. The implemented experiments show it performs well in the process of data swapping, inquiring, and dynamic updating for massive multi-source data under urban-block scale.
出处
《遥感信息》
CSCD
北大核心
2017年第5期15-22,共8页
Remote Sensing Information
基金
国家自然科学基金(41401527)
测绘地理信息公益性行业科研专项(201512008)
武汉市测绘研究院博士后创新实践基地科研项目(WGF2016001)
关键词
车载移动扫描
多源海量数据
管理与动态调度
瓦片索引
随机采样八叉树
vehicle-borne mobile scanning
massive multi-source data
organization and dynamic swapping
tile indexing
random sampling octree
作者简介
谢洪（1987-），男，博士研究生，主要研究方向为三维激光测量技术及应用。E-mail：hxie@sgg．whu.edu．cn
