邻轨PS-InSAR地面沉降结果拼接处理方法与实验
|
摘要
目前,我国许多经济发达地区都面临着地面沉降灾害的困扰,沉降范围扩大,程度日益加剧,逐渐成为城市发展中亟待解决的问题。永久散射体合成孔径雷达干涉测量(Persistent Scatterer Interferometric SyntheticAperture Radar,PS-InSAR)作为地表形变测量的主要手段之一,在地面沉降监测中发挥着重要作用。而单轨星载SAR影像成像幅宽有限,在开展大范围地面沉降调查时需要将多轨道PS-InSAR沉降速率图进行拼接。本文重点讨论了入射角效应和参考点差异对PS-InSAR沉降结果的影响,分析了相邻轨道PS-InSAR沉降速率拼接中存在的PS点位置差异和沉降量偏移,鉴此,提出了采用区块法和插值法对异轨重叠区的形变结果求差的思路,以及基于现有软件PS-InSAR地面沉降速率的跨轨拼接处理流程,利用广东珠三角地区ENVISAT ASAR数据进行了实验分析。结果表明,相邻轨道入射角不同会造成沉降量的差异,在多轨道情况下对沉降量影响增大,因此,在拼接过程中需要进行入射角纠正。本文提出的区块法和插值法能有效地求解异轨重叠区的形变差,结果表明区块法优于插值法;相邻轨道参考点差异会造成沉降量偏移,通过区块法或插值法求差可以消除该偏移量。本文提出的拼接流程可将多轨道PS-InSAR地面沉降速率统一到同一基准下,从而获得大范围一致的地面沉降速率。
At present, more and more economic developed areas are facing serious land subsidence problems with increasing coverage and intensity, which prevents the application of sustainable development in these areas.Thus, the land subsidence problems should be controlled urgently. PS-InSAR technology, as one of the most important tools to detect land subsidence, has played a significant role in land subsidence monitoring. However,multi-track PS-InSAR results must be connected to obtain the large scale subsidence map when an extended land subsidence survey area is required, since spaceborne SAR image from only one track is too limited in scan range to cover the whole area. This paper analyzes the processing chain of PS-InSAR and points out two key problems when connecting multi-track PS-InSAR results from adjacent tracks. First, many PS from adjacent tracks are physically different, which is mainly caused by different incidence angles and coherences of adjacent tracks.Therefore, it is hard to find the tie point in the overlapped area. Second, there is a subsidence offset between two adjacent tracks according to the overlapped area, which is attributed to the physical differences of PS and the processing parameters of PS-InSAR, such as the reference point, the temporal coverage and so on. Among these factors, impacts caused by different incidence angles and reference offset are emphasized. Two new methods, Block Method and Interpolation Method, are proposed in this paper to calculate the differences between subsidence velocities from adjacent tracks caused by the physical differences of PS in the overlapped area. This can avoid the difficulty of finding tie point in the overlapped area. Moreover, subsidence velocities are adjusted to mitigate the impacts from different incidence angles. Finally, based on the existing software, a complete procedure for multitrack PS-InSAR subsidence results connection is demonstrated. At the end of this paper, ENVISAT ASAR datasets of Pearl Delta River in Guangdong province are processed and analyzed in experiments. Experimental results show that different incidence angles could affect the land subsidence from adjacent tracks, so it is necessary to correct the differential subsidence caused by the incidence difference, especially when multiple tracks are involved. Block Method and Interpolation Method are all effective in calculating the differences of subsidence velocities, and Block Method was superior to Interpolation Method through statistical analyses. Meanwhile, different incidence angles and reference offset are effectively mitigated to obtain the conformed subsidence velocities using these two methods. Consequently, PS-InSAR subsidence velocities from two ENVISAT tracks are connected effectively, which shows the effectiveness of the processing procedure proposed in this paper.
At present, more and more economic developed areas are facing serious land subsidence problems with increasing coverage and intensity, which prevents the application of sustainable development in these areas.Thus, the land subsidence problems should be controlled urgently. PS-InSAR technology, as one of the most important tools to detect land subsidence, has played a significant role in land subsidence monitoring. However,multi-track PS-InSAR results must be connected to obtain the large scale subsidence map when an extended land subsidence survey area is required, since spaceborne SAR image from only one track is too limited in scan range to cover the whole area. This paper analyzes the processing chain of PS-InSAR and points out two key problems when connecting multi-track PS-InSAR results from adjacent tracks. First, many PS from adjacent tracks are physically different, which is mainly caused by different incidence angles and coherences of adjacent tracks.Therefore, it is hard to find the tie point in the overlapped area. Second, there is a subsidence offset between two adjacent tracks according to the overlapped area, which is attributed to the physical differences of PS and the processing parameters of PS-InSAR, such as the reference point, the temporal coverage and so on. Among these factors, impacts caused by different incidence angles and reference offset are emphasized. Two new methods, Block Method and Interpolation Method, are proposed in this paper to calculate the differences between subsidence velocities from adjacent tracks caused by the physical differences of PS in the overlapped area. This can avoid the difficulty of finding tie point in the overlapped area. Moreover, subsidence velocities are adjusted to mitigate the impacts from different incidence angles. Finally, based on the existing software, a complete procedure for multitrack PS-InSAR subsidence results connection is demonstrated. At the end of this paper, ENVISAT ASAR datasets of Pearl Delta River in Guangdong province are processed and analyzed in experiments. Experimental results show that different incidence angles could affect the land subsidence from adjacent tracks, so it is necessary to correct the differential subsidence caused by the incidence difference, especially when multiple tracks are involved. Block Method and Interpolation Method are all effective in calculating the differences of subsidence velocities, and Block Method was superior to Interpolation Method through statistical analyses. Meanwhile, different incidence angles and reference offset are effectively mitigated to obtain the conformed subsidence velocities using these two methods. Consequently, PS-InSAR subsidence velocities from two ENVISAT tracks are connected effectively, which shows the effectiveness of the processing procedure proposed in this paper.
引文
[1]林珲,陈富龙,江利明,等.多基线差分雷达干涉测量的大型人工线状地物形变监测[J].地球信息科学学报,2010,12(5):718-724.
[2]Ketelaar G,Leijen F V,Marinkovic P,et al.Multi-track PS-InSAR datum connection[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2007:2481-2484.
[3]Ketelaar G,Leijen F V,Marinkovic,et al.Multi-track PSInSAR:Datum connection and reiliability assessment[C].Proceedings of Envisat Symposium,2007.
[4]Perissin D,Prati C,Rocca F.ASAR parallel-track PS analysis in urban sites[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2007:1167-1170.
[5]Perissin D,Prati C,Engdahl M E,et al.Validating the SAR wave number shift principle with the ERS-Envisat PS coherent combination[J].IEEE transactions on Geoscience and Remote Sensing,2006,44(9):2343-2351.
[6]Perissin D,Prati C,Engdahl M E,et al.Multi-track PS analysis in Shanghai[C].Proceedings of Envisat Symposium,2007:23-27.
[7]Ge D,Wang Y,Zhang L,et al.Large scale land subsidence monitoring with a reduced set of SAR images[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2009:558-561.
[8]Ge D,Zang L,Wang Y,et al.Merging multi-track PSI result for land subsidence mapping over very extended area[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2010:3522-3525.
[9]Ge D,Wang Y,Zang L,et al.Using Permanent Scatterer InSAR to monitor land subsidence along high speed railway-The first experiment in China[C].Proceedings of Fringe,2009.
[10]Taku O,Hideki U.Advanced Interferometric Synthetic Aperture Radar(InSAR)time series analysis using interferograms of multiple-orbit tracks:A case study on Miyake-jima[J].Journal of Geophysical Research,2011,116(1):1-14.
[11]崔喜爱,曾琪明,焦健,等.大气效应对合成孔径雷达干涉测量的影响[J].中国科技论文,2013,8(4):302-306.
[12]Cui X A,Zeng Q M,Liang C R,et al.Research of InSAR atmospheric correction using WRF model simulated result and MERIS product[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2012:3627-3630.
[13]张玉玺,孙继朝,陈玺,等.珠江三角洲浅层地下水pH值的分布及成因浅析[J].水文地质工程地质,2011,38(1):16-20.
[14]林本海,杨树庄,朱伯善,等.广东省地质构造与岩土工程基本特征[J].岩石力学与工程学报,2006,25(2):3337-3345.
[15]汤连生,廖化荣,廖志强,等.珠江三角洲环境地质分区及其特征[J].中山大学学报(自然科学版),2004,43(1):229-232.
[16]Zhao Q,Lin H,Jiang L.Ground deformation monitoring in Pearl River Delta Region with stacking D-InSAR technique[C].Proceedings of SPIE 7145,Geoinformatics2008 and Joint Conference on GIS and Built Environment:Monitoring and Assessment of Natural Resources and Environments,2008.
[2]Ketelaar G,Leijen F V,Marinkovic P,et al.Multi-track PS-InSAR datum connection[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2007:2481-2484.
[3]Ketelaar G,Leijen F V,Marinkovic,et al.Multi-track PSInSAR:Datum connection and reiliability assessment[C].Proceedings of Envisat Symposium,2007.
[4]Perissin D,Prati C,Rocca F.ASAR parallel-track PS analysis in urban sites[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2007:1167-1170.
[5]Perissin D,Prati C,Engdahl M E,et al.Validating the SAR wave number shift principle with the ERS-Envisat PS coherent combination[J].IEEE transactions on Geoscience and Remote Sensing,2006,44(9):2343-2351.
[6]Perissin D,Prati C,Engdahl M E,et al.Multi-track PS analysis in Shanghai[C].Proceedings of Envisat Symposium,2007:23-27.
[7]Ge D,Wang Y,Zhang L,et al.Large scale land subsidence monitoring with a reduced set of SAR images[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2009:558-561.
[8]Ge D,Zang L,Wang Y,et al.Merging multi-track PSI result for land subsidence mapping over very extended area[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2010:3522-3525.
[9]Ge D,Wang Y,Zang L,et al.Using Permanent Scatterer InSAR to monitor land subsidence along high speed railway-The first experiment in China[C].Proceedings of Fringe,2009.
[10]Taku O,Hideki U.Advanced Interferometric Synthetic Aperture Radar(InSAR)time series analysis using interferograms of multiple-orbit tracks:A case study on Miyake-jima[J].Journal of Geophysical Research,2011,116(1):1-14.
[11]崔喜爱,曾琪明,焦健,等.大气效应对合成孔径雷达干涉测量的影响[J].中国科技论文,2013,8(4):302-306.
[12]Cui X A,Zeng Q M,Liang C R,et al.Research of InSAR atmospheric correction using WRF model simulated result and MERIS product[C].Proceedings of IEEE International Geoscience and Remote Sensing Symposium,2012:3627-3630.
[13]张玉玺,孙继朝,陈玺,等.珠江三角洲浅层地下水pH值的分布及成因浅析[J].水文地质工程地质,2011,38(1):16-20.
[14]林本海,杨树庄,朱伯善,等.广东省地质构造与岩土工程基本特征[J].岩石力学与工程学报,2006,25(2):3337-3345.
[15]汤连生,廖化荣,廖志强,等.珠江三角洲环境地质分区及其特征[J].中山大学学报(自然科学版),2004,43(1):229-232.
[16]Zhao Q,Lin H,Jiang L.Ground deformation monitoring in Pearl River Delta Region with stacking D-InSAR technique[C].Proceedings of SPIE 7145,Geoinformatics2008 and Joint Conference on GIS and Built Environment:Monitoring and Assessment of Natural Resources and Environments,2008.