Sentinel-1卫星TOPS模式影像增强谱分集配准优化  

Advanced ESD coregistration of inteferometric processing for Sentinel-1 TOPS data

作　　者：吴文豪 张磊[2] 张腾旭 王明洲 龙四春 段梦[5] 周志伟 祝传广 WU Wenhao;ZHANG Lei;ZHANG Tengxu;WANG Mingzhou;LONG Sichun;DUAN Meng;ZHOU Zhiwei;ZHU Chuanguang(Hunan Province Key Laboratory of Coal Resources Clean-utilization and Mine Environment Protection,Hunan University of Science and Technology,Xiangtan 411201,China;Department of Land Surveying and Geo-informatics,the Hong Kong Polytechnic University,Hongkong 999077,China;School of Resources and Environment Science and Engineering,Hubei University of Science and technology,Xianning 437100,China;Shenzhen Water Planning and Design Institute Company Limited,Shenzhen 518000,China;School of Geosciences and Info-physics,Central South University,Changsha 410083,China;Innovation Academy for Precision Measurement Science and Technology,Chinese Academy of Sciences,Wuhan 430077,China)

机构地区：[1]湖南科技大学煤炭资源清洁利用与矿山环境保护湖南省重点实验室,湖南湘潭411201 [2]香港理工大学土地测量及地理资讯学系,中国香港999077 [3]湖北科技学院资源环境科学与工程学院,湖北咸宁437100 [4]深圳市水务规划设计院股份有限公司,广东深圳518000 [5]中南大学地球科学与信息物理学院,湖南长沙410083 [6]中国科学院精密测量科学与技术创新研究院,湖北武汉430077

出　　处：《测绘学报》2020年第11期1451-1462,共12页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金(41877283);湖南省自然资源调查与监测工程技术研究中心开放课题(2020-16);湖南省教育厅科学研究项目(18C0316)。

摘　　要：Sentinel-1卫星TOPS模式影像通常在几何配准的基础上,再次利用Burst重叠区域进行增强谱分集处理以实现高精度配准。几何配准主要依赖卫星轨道参数,难以进一步提高其配准精度,因此Sentinel-1影像配准的关键则是通过增强谱分集准确估计出几何配准后的残余偏移量。然而,增强谱分集易受失相干噪声等因素的影响,低相干条件下难以满足0.001像素的配准精度要求。因此本文首先针对单基线条件下的增强谱分集配准处理进行以下改进:①完善增强谱分集多视理论,采用前置多视处理,优化增强谱分集处理流程;②基于增强谱分集相位的残余偏移量等权估计升级为加权估计,改善参数估计方法;③完善距离向增强谱分集加权估计理论,实施距离向增强谱分集配准,增加增强谱分集的观测量。然后在单基线配准的基础上,针对时序影像处理提出优化方案:①根据影像干涉对组合所形成的增强谱分集冗余观测量进行多基线配准;②采用分布式目标技术提高增强谱分集干涉图相位质量。上述改进方案互相补充,不但能独立实施,而且可以联合使用。研究结果表明,以上改进方法能够在一定程度上提高配准精度。The TOPS mode images of Sentinel-1 satellite usually use the overlapping regions between bursts to achieve high-precision coregistration by using enhanced spectral diversity(ESD)technique after carrying out geometric coregistration.The geometric coregistration relies on the use of satellite orbit parameters which limiting the registration accuracy.Consequently,the key of sentinel-1 image registration is to accurately estimate the residual offset after geometric registration through enhanced spectral diversity.However,the ESD suffers from incoherent noise making it difficult to meet the requirement of 0.001 of the pixels spacing.In this study,we first improve the ESD performance for single interferometric pair in following aspects:①develop ESD multi-looking theory,optimized the ESD workflow by using early-multi-looking process;②apply weighted periodogram to improve the parameter estimation on the basis of the residual offset equal weight observations of ESD phase;③propose the ESD weighted estimation that exploits range overlapping areas,and an increased number of overlapping areas contribute to the estimated accuracy.On this basis of single-baseline co-registration,we proposed the optimized solutions for time-series image processing:①use the redundant ESD observations from interferometric pairs to conduct multi-baseline coregistration;②improve the interferogram quality by distributed scatterers technique so that increase the accuracy of the ESD estimation.The above improvement schemes are mutually complemental and can also be implemented independently.The experiment results show that the above improvement methods can improve the registration accuracy.

关 键 词：哨兵卫星 增强谱分集 周期图谱 分布式目标 前置多视 后置多视 

分 类 号：P237[天文地球—摄影测量与遥感]