首页 | 官方网站   微博 | 高级检索  
     

海岛礁卫星影像无地面控制点定位方法
引用本文:陈小卫,郭海涛,张保明,岑敏仪,卢俊.海岛礁卫星影像无地面控制点定位方法[J].武汉大学学报(信息科学版),2019,44(6):933-940.
作者姓名:陈小卫  郭海涛  张保明  岑敏仪  卢俊
作者单位:1.北京跟踪与通信技术研究所, 北京, 100094
基金项目:国家自然科学基金41601507地理信息工程国家重点实验室开放基金SKLGIE2015-M-3-3
摘    要:针对海岛礁卫星影像的定位问题,提出一种利用航天飞机雷达地形测绘任务(shuttle radar topography mission,SRTM)DEM辅助的无地面控制点定位方法。该方法分为概略定位和精定位两个阶段,各阶段均包括DEM提取和DEM匹配等主要步骤,可分别对影像中的相对误差和绝对误差进行补偿。SRTM DEM被充分应用到方法各环节中,以发挥其高精度的特性:提取DEM时既用于剔除海域点,也用于确定求解陆域点高程时的高程搜索范围,从而避免海域影像的不利影响,同时保证计算效率;DEM匹配时其作为基准数据。利用多景天绘一号卫星海岛礁地区的立体影像进行验证。实验结果表明,所提出的方法对具有不同陆域比例、不同生产方式的天绘一号海岛礁影像均能得到较稳定且较高的定位精度,平面和高程精度分别优于6.2 m、5.2 m,能较好地满足1:50 000比例尺地形图的精度要求。定位精度基本不受待匹配DEM分辨率的影响,计算效率取决于陆域比例和待匹配DEM的分辨率。

关 键 词:海岛礁影像    无地面控制点定位    SRTM  DEM    自由网平差    DEM匹配
收稿时间:2018-06-10

Geopositioning Without Ground Control Points for Satellite Image of Island and Reef
Affiliation:1.Beijing Institute of Tracking and Telecommunications Technology, Beijing 100094, China2.Institute of Surveying and Mapping, Information Engineering University, Zhengzhou 450001, China3.Faculty of Geosciences and Environmental Engineering, Southwest Jiaotong University, Chengdu 611756, China
Abstract:A shuttle radar topography mission (SRTM) DEM assistant geopositioning method is proposed to solve the problem of geopositioning without ground control points (GCPs) for satellite island and reef images. The method is divided into two stages:coarse and accurate correction. DEM extraction and DEM matching are key steps for both stages, which aim at compensating the relative and absolute errors in triplet image, respectively. In order to fully take advantage of its consistent and stable high precision characteristic of it, SRTM DEM is integrated into the method in each part. In DEM extraction step, it is used to eliminate the sea area and also to determine the vertical search range for solving the land area elevation, so as to avoid the adverse effect of the sea area images and ensure computational efficiency at the same time. As for DEM matching, it is regarded as reference data. Many scenes of Mapping Satellite-1 island and reef images are used to validate the performance of this method. Experimental results show that the method can obtain stable and high geopositioning accuracy for Mapping Satellite-1 island and reef images, regardless of land area proportion or production mode. The planimetric and vertical accuracy are better than 6.2 m and 5.2 m respectively, which can satisfy the accuracy requirements of 1:50 000 scale mapping. The geopositioning accuracy is basically not affected by target DEM resolution, and the computational efficiency depends on proportion of land area and target DEM resolution.
Keywords:
本文献已被 CNKI 等数据库收录!
点击此处可从《武汉大学学报(信息科学版)》浏览原始摘要信息
点击此处可从《武汉大学学报(信息科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号