融合DT和SDFPT的时序InSAR矿区形变监测与分析  

Deformation monitoring and analysis of mining areas based on the DT-SDFPT combined time-series InSAR

作　　者：于冰[1,2,3,4] 王冰 刘国祥 张过 胡云亮[1] 胡金龙 YU Bing;WANG Bing;LIU Guoxiang;ZHANG Guo;HU Yunliang;HU Jinlong(School of Civil Engineering and Geomatics,Southwest Petroleum University,Chengdu 610500,China;Institude of Petroleum and Natural Gas Spatial Information Engineering,Southwest Petroleum University,Chengdu 610500,China;State Key Laboratory of Information Engineering in Surveying,Mapping&Remote Sensing,Wuhan University,Wuhan 430079,China;State Key Laboratory of Geodesy and Earth’s Dynamics,Innovation Academy for Precision Measurement Science and Technology,Chinese Academy of Sciences,Wuhan 430077,China;Faculty of Geosciences and Environmental Engineering,Southwest Jiaotong University,Chengdu 611756,China)

机构地区：[1]西南石油大学土木工程与测绘学院,成都610500 [2]西南石油大学油气空间信息工程研究所,成都610500 [3]武汉大学测绘遥感信息工程国家重点实验室,武汉430079 [4]中国科学院精密测量科学与技术创新研究院大地测量与地球动力学国家重点实验室,武汉430077 [5]西南交通大学地球科学与环境工程学院,成都611756

出　　处：《自然资源遥感》2024年第1期14-25,共12页Remote Sensing for Natural Resources

基　　金：国家自然科学基金项目“基于卫星升降轨时序DInSAR的塔里木油田沉降监测及储层状态参数反演”(编号:41801399);测绘遥感信息工程国家重点实验室开放基金项目“基于星载SAR干涉的克拉玛依油田时序二维形变监测及储层参数反演”(编号:18E01);大地测量与地球动力学国家重点实验室开放基金项目“玛湖特大油田InSAR沉降监测及储层动力学参数反演”(编号:SKLGED2020-5-1-E);四川省杰出青年科技人才项目“西南地区植被干旱胁迫遥感监测与预警”(编号:2021JDJQ0007)共同资助。

摘　　要：煤矿高强度开采会导致严重的地面形变及次生地质灾害。时序合成孔径雷达干涉(synthetic aperture Radar interferometry,InSAR)具有较强的形变监测能力,但在开采核心及周边低相干区域时无法监测到足够的目标点。该文尝试将分布式目标(distributed target,DT)和缓慢去相关滤波相位目标(slowly-decorrelating filtered phase target,SDFPT)进行联合,以提高矿区形变监测点的密度和覆盖度。分别采用快速同质点选取(fast statistically homogenous pixel selection,FaSHPS)法和振幅离差指数法选取DT和SDFPT候选点,分别对2类点进行相位优化和稳定性分析,筛选出符合条件的DT和SDFPT形成融合点集,并对其进行三维相位解缠、恢复相位时间序列和时空滤波,最终得到融合点集的形变时间序列和年均形变速率。选取2018年4月—2020年4月获取的覆盖布尔台煤矿的60景Sentinel-1影像进行形变监测,结果表明,融合DT和SDFPT后形变点密度和覆盖度显著提升,可监测最大形变量级也随之增加。实验区域内存在5处形变漏斗,最大累积形变量达到-309.76 mm;形变影响范围和不同年份时序形变量的差异与矿区开采活动密切相关。High-intensity coal mining leads to significant surface deformation and secondary geological disasters.Synthetic aperture Radar interferometry(InSAR),exhibiting high deformation monitoring capability,fails to detect enough target pixels in the mining core and surrounding low-coherence areas.This study intends to increase the density and coverage of deformation monitoring points in mining areas by combining distributed targets(DTs)and slowly-decorrelating filtered phase targets(SDFPTs).First,DT and SDFPT candidate pixels were selected using the fast statistically homogenous pixel selection(FaSHPS)method and the amplitude dispersion index method,respectively for phase optimization and stability analysis.Then,qualified DT and SDFPT pixels were screened out to constitute a fused pixel set,which was subjected to three-dimensional phase unwrapping,phase time series recovery,and spatio-temporal filtering.Consequently,the deformation time series and the annual average deformation rate were determined based on the fused pixel set.Finally,the method proposed in this study was applied to monitor the deformation in the Buertai coal mine using 60 scenes of Sentinel-1 images covering the coal mine from April 2018 to April 2020.The results reveal a significant increase in the density and coverage of deformation points through the integration of DT and SDFPT,thus allowing for the monitoring of higher levels of maximum deformation.Within the experimental area,five deformation cones were identified,with the maximum cumulative deformation amplitude reaching-309.76 mm.The influencing range of the deformations and the difference in the deformation amplitude of the time series in different years are closely related to mining activities.

关 键 词：布尔台煤矿 分布式目标 缓慢去相关滤波相位目标 地表形变 Sentinel-1A 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]