机载阵列SAR点云区域网平差  

Block Adjustment of Airborne Array SAR Point Cloud

作　　者：潘凤卓 程春泉 黄国满[2] 纪方方[2,3] PAN Fengzhuo;CHENG Chunquan;HUANG Guoman;JI Fangfang(College of Geodesy and Geomatics,Shandong University of Science and Technology,Qingdao,Shandong 266590,China;Chinese Academy of Surveying and Mapping,Beijing 100036,China;College of Resource Environment and Tourism,Capital Normal University,Beijing 100048,China)

机构地区：[1]山东科技大学测绘与空间信息学院,山东青岛266590 [2]中国测绘科学研究院,北京100036 [3]首都师范大学资源环境与旅游学院,北京100048

出　　处：《遥感信息》2024年第3期113-120,共8页Remote Sensing Information

基　　金：国家重点研发计划(2022YFB3902601、2022YFB3901604、2022YFB3902605)。

摘　　要：阵列SAR技术作为一种新型的三维重建技术,是InSAR发展一个重要的方向。由于其存在与点云位置成非线性关系的高频误差,难以直接用于测绘产品生产。针对阵列SAR点云中误差的特点,首先对点云进行了噪点去除和点云匹配,然后对平面坐标采用线性精化模型,在对高程坐标采用二次多项式精化模型进行平差的基础上,通过增加对虚拟观测值的约束,构建阵列SAR点云数据的区域网平差模型。最后选择山区和城区两种典型地表数据进行实验。分析结果表明:文章所构建的区域网平差模型可有效降低阵列SAR点云定位中存在的高频误差影响;在不同控制点情况下,定位精度和解算稳健性均得到有效提升。Array SAR technology,as a new type of 3D reconstruction technology,is an important direction for the development of InSAR.Due to the presence of high-frequency errors that are non-linear to the position of the point cloud,it is difficult to directly apply them to the production of surveying and mapping products.This article focuses on the characteristics of errors in array SAR point clouds.Firstly,noise is removed and point cloud matching is performed on the point cloud.Then,a linear refinement model is used for plane coordinates and a quadratic polynomial refinement model is used for elevation coordinate adjustment.By adding constraints on virtual observation values,a block adjustment model for array SAR point cloud data is constructed.Finally,two typical surface data types,which are mountainous and urban areas,are selected for experiments.The results show that the block adjustment model constructed in this paper can effectively reduce the high-frequency error impact in array SAR point cloud positioning,and the positioning accuracy and computational robustness are effectively improved under different control point conditions.

关 键 词：阵列SAR 点云滤波 点云匹配 区域网平差 RANSAC 

分 类 号：P231.5[天文地球—摄影测量与遥感]