InSAR时序分析高相干目标选取方法比较研究

doi:10.3724/SP.J.1047.2016.00805

摘要/Abstract

摘要：

合成孔径雷达干涉测量（Interferometric Synthetic Aperture Radar,InSAR）时序分析技术,利用时域上多幅SAR图像选择时间基线、空间基线满足一定条件的干涉对进行干涉,通过建立干涉相位形变模型获取地表形变信息。InSAR时序分析技术改善了差分干涉测量中时空失相干、大气延迟等问题,广泛应用于有关形变监测的多个领域,并逐渐成为获取长期地表形变趋势的重要手段。在InSAR时序分析中,针对不同应用选取合适的高相干目标选取方法,获得可靠的高相干目标,是获取精确可靠的地表形变信息的基础。本文通过分析永久散射体（Permanent Scatterers,PS）、分布式散射体（Distributed Scatterers,DS）选取方法的理论模型及其算法,研究其适用地物目标类型的异同,并分析、归纳总结了幅度相关法、相位分析法、信号杂波比法、相干性统计法等不同的高相干目标选取方法的优缺点。最后,以阿尔金断裂带西段部分区域为研究区,分别采用具有代表性的PS、DS选取方法开展该研究区域的选点实验,结果表明该研究区域DS选取方法比PS选取方法适用。本文方法为解决在不同地理区域进行应用研究时选取合适的选点方法提供参考。

关键词: InSAR时序分析, PS, DS

Abstract:

Synthetic Aperture Radar Interferometry (InSAR) time series analysis technique gradually becomes an important approach to obtain the long-term trend of surface deformation. It chooses a series of SAR images and establishes the corresponding model to extract the surface deformation. Currently, the InSAR time analysis technique is capable to improve the decorrelation and atmospheric delay in InSAR, therefore it has been widely used in many fields, including the urban ground deformation caused by the exploitation of underground resources, the mining field surface deformation caused by exploitation of mineral and natural gas, and the natural surface deformation caused by earthquake, volcano or landslide. Among different practical cases, selecting an appropriate method to capture the high coherent target is the fundamental step in getting the accurate and reliable surface deformation information. In this paper, we analyze the theoretical models of Permanent Scatterers (PS) and Distributed Scatterers (DS) and the differences of characteristics between the PS and DS point types. And we summarize the advantages and disadvantages of different high coherent target selection methods from the perspectives of SAR echo signal and InSAR time series analysis. Dispersion of amplitude is the most widely used method and the phase analysis method is more suitable for further selecting the points of PS. For the method of signal-to-clutter ratio, it is more applicable to select the PS candidates. The correlation coefficient method is only adaptive to a few SAR data. And for DS selection, the improved average temporal coherence method is more representative than the dispersion of amplitude difference method and the average temporal coherence method. In addition, this paper chooses some of the western part of Altyn Tagh Fault as the study area. And we verify the feasibility of the typical PS and DS selection methods to the real surface of the study area. The experimental results show that the DS method is more applicable than the PS method in non-urban area, and we obtain the deformation rate of Altyn Tagh Fault as well. Through these analyses, we provide the theory basis for selecting the appropriate methods in different research applications within different geographical areas.

Key words: InSAR time-series analysis, PS, DS

范锐彦, 焦健, 高胜, 曾琪明. InSAR时序分析高相干目标选取方法比较研究[J]. 地球信息科学学报, 2016, 18(6): 805-813.DOI:10.3724/SP.J.1047.2016.00805

FAN Ruiyan,JIAO Jian,GAO Sheng,ZENG Qiming. Comparison Research of High Coherent Target Selection Based on InSAR Time Series Analysis[J]. Journal of Geo-information Science, 2016, 18(6): 805-813.DOI:10.3724/SP.J.1047.2016.00805

图/表 9

图1

表1

高相干目标选取方法比较

目标选取方法		公式	辐射校正	SAR影像数量要求	优点	缺点	应用范围
点目标选取方法	振幅离差法	$D A = σ A m A$	是	一般要有25~30景以上SAR数据支持	所选目标质量较高,运算效率高	未考虑目标本身的幅度值,容易造成误判	适合城市区域
	强度离差法	$D I = σ I m I$	否	数量要求较多	阈值更加宽泛,PS点质量更高	与DA法差异不大,无明显提高
	相位分析法	$γ x = 1 N \| exp - 1 {? k x - ? ? k x - ? angl, k (x)}) \|$	否	数量要求较多	提取的PS点分布比较均匀	但 $γ$ 的计算过程步骤较为复杂,运算量很大
	信号杂波比法	$SCR = I S I N$	否	数量要求较多	不需要辐射校正	估计的窗口大小会对结果产生较大的影响
	子视相关系数法	$ρ x, y = ∑ s 1 (x, y) s 2 * (x, y) ∑ \| s 1 x, y \| 2 ∑ \| s 2 * x, y \| 2$	否	数量要求较少	适合于SAR影像数量较少时PS点目标的选取	运算量大,筛选效率低;移动窗口的大小不好把握
面目标选取方法	幅度差离差法	$σ φ ≈ σ ? A m A$	是	数量要求较多	适合短时空基线的相位较为稳定,后向散射并不是那么强的目标	并未考虑其目标是统计分布特征	适合非城区
	时序平均相干性法	$ρ ? x, y = 1 N ∑ i = 1 N ρ i (x, y)$	否	数量要求较少	模型简单,运算效率高	移动窗口大小不好把握
	改进的时序平均相干性法	$D N = max - ∝ < x < ∝ S N (x) - P N (x)$	否	数量要求较多	将面状目标转化为同质像元,相位估计更加准确	移动窗口大小不好把握

表1

图2

图3

表2

表3

图4

图5

图6

参考文献 19

[1]	Bamler R, Hartl P.Synthetic aperture radar interferometry[J]. Inverse Problems, 1998,14(4):R1-R54.
[2]	Arnaud A, Closa J, Hanssen R, et al.Development of algorithms for the exploitation of ERS-Envisat using the stable points network[Z]. Altamira Information, Barcelona, Spain, 2004.
[3]	Ferretti A, Prati C, Rocca F.Permanent scatterers in SAR interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2001,39(1):8-20.
[4]	徐小波,屈春燕,单新建,等.基于PS-InSAR技术的断裂带地壳形变实验研究[J].地球科学进展,2012(4):452-459.
	[Xu X B, Qu C Y, Shan X J, et al.An experimental study of monitoring fault crustal deformation using PS-InSAR technology[J]. Advances in Earth Science, 2012,4:452-459. ]
[5]	Berardino P, Fornaro G, Lanari R, et al.A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002,40(11):2375-2383.
[6]	高胜. 联合PS和DS的InSAR时间序列分析方法研究[D].北京:北京大学,2015.
	[Gao S.A method of combining PS and DS for InSAR time series analysis[D]. Beijing: Peking University, 2015 ]
[7]	Ferretti A, Fumagalli A, Novali F, et al.A new algorithm for processing interferometric data-stacks: SqueeSAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011,49(9):3460-3470. doi: 10.1109/TGRS.2011.2124465
[8]	Ferretti A, Prati C, Rocca F.Analysis of permanent scatterers in SAR interferometry[J]. IGARSS, 2000,2:761-763.
[9]	罗小军,黄丁发,刘国祥.时序差分雷达干涉中永久散射体的自动探测[J].西南交通大学学报,2007,42(4):414-418.
	[Luo X J, Huang D F, Liu G X.Automated detection of permanent scatterers in time serial differential radar interferometry[J]. Journal of Southwest Jiaotong University, 2007,42(4):414-418. ]
[10]	Hooper A.A multi-temporal InSAR method incorporating both persistent scatterer and small baseline approaches[J]. Geophysical Research Letters, 2008,35(16):L16302.
[11]	Hooper A, Segall P, Zebker H.Persistent scatterer interferometric synthetic aperture radar for crustal deformation analysis, with application to Volcán Alcedo, Galápagos[J]. Journal of Geophysical Research Atmospheres, 2007, 112(B7): B07407.
[12]	Ainsworth T L, Jansen R W, Lee J S, et al.Sub-aperture analysis of high-resolution polarimetric SAR data[C]. Geoscience and Remote Sensing Symposium, 1999,1:41-43.
[13]	陶秋香. PS InSAR关键技术及其在矿区地面沉降监测中的应用研究[D].青岛:山东科技大学,2009.
	[Tao Q X.Study of key techniques of PS InSAR and its applications to monitor mining land subsidence[D]. Qingdao: Shandong University of Science and Technology, 2009. ]
[14]	Mora O, Lanari R, Mallorquí J J, et al.A new algorithm for monitoring localized deformation phenomena based on small baseline differential SAR interferograms[C]. Geoscience and Remote Sensing Symposium, 2002,2:1237-1239.
[15]	陈强,刘国祥,李永树,等.干涉雷达永久散射体自动探测——算法与实验结果[J].测绘学报,2006(2):112-117.
	[Chen Q, Liu G X, Li Y S, et al.Automated detection of permanent scatterers in Radar interferometry: algorithm and testing results[J]. Acta Geodaetica et Cartographica Sinica, 2006,2:112-117. ]
[16]	国家地震局阿尔金活动断裂课题组.阿尔金活动断裂带[M].北京:地震出版社,1992.
	[ The Research Group of the Altun Fault Zone, State Seismological Bureau. The active Altun Fault zone[M]. Beijing: Sesmo Logical Press,1992. ]
[17]	Wallace K, Yin G, Bilham R.Inescapable slow slip on the AltynTagh Fault[J]. Geophysical. Research Letters, 2004,31: L09613.
[18]	Zhang P Z, Burchfiel B C, Molnar P, et a1. Amount and style of late Cenozoic deformation in the LiupanShan area, Ningxia Autonomous Region, China[J]. Tectonics, 1991,10(6):1111-1129.
[19]	Liang S, Gan W, Shen C, et al.Three-dimensional velocity field of present-day crustal motion of the Tibetan Plateau derived from GPS measurements[J]. Journal of Geophysical Research, 2013,118:1-11.

序号	主影像获取日期	辅影像获取日期	垂直基线 B_⊥/m	时间基线/d
1	2009-01-06	2006-10-24	-440.4	805
2	2009-01-06	2006-11-28	-77.6	770
3	2009-01-06	2007-02-06	-218.5	700
4	2009-01-06	2007-03-13	344.3	665
5	2009-01-06	2007-07-31	-100.2	525
6	2009-01-06	2007-11-13	219.8	420
7	2009-01-06	2008-05-06	-54.3	245
8	2009-01-06	2008-06-10	104.7	210
9	2009-01-06	2008-07-15	140.1	175
10	2009-01-06	2008-09-23	-226.7	105
11	2009-01-06	2008-10-28	19.6	70
12	2009-01-06	2009-02-10	-184.8	35
13	2009-01-06	2009-03-17	488.0	70
14	2009-01-06	2009-04-21	-164.9	105
15	2009-01-06	2009-05-26	74.4	140
16	2009-01-06	2009-08-04	-116.6	210
17	2009-01-06	2009-10-13	-256.7	280
18	2009-01-06	2009-12-22	-191.9	350
19	2009-01-06	2010-01-26	177.4	385
20	2009-01-06	2010-03-02	-194.4	420
21	2009-01-06	2010-04-06	307.0	455
22	2009-01-06	2010-06-15	187.3	525

选取方法	名称	数量/个	密度/（个/km²）
PS选点方法	DA	124 125	56.42
PS选点方法	相位分析法	107 355	48.80
DS选点方法	DAD	3 408 784	1549.45
DS选点方法	改进的时序平均相干性法	3 635 002	1652.27