滑坡灾害InSAR应急排查技术方法研究

研究探索了一条以InSAR变形为主,辅助光学遥感、地貌和地质条件特征,进行活动性滑坡快速排查的技术方法,并以四川省雷波县域为例进行了实验。应用4种SAR数据（PALSAR-1升轨、Sentinel-1 A/B升、降轨和PALSAR-2降轨数据）进行合成孔径雷达干涉（Interferometric Synthetic Aperture Radar,InSAR）处理,共解译活动性地质灾害163处,并分析了其时空分布规律,获得了几点认识:（1）InSAR技术能够追溯滑坡长期微小变形,从而实现活动性滑坡的有效识别;（2）多时段、多角度和多分辨率SAR数据的综合使用,可以有效克服滑坡观测的阴影叠掩、失相干等问题,提高滑坡（尤其是高位滑坡）识别的效果、效率和时效性;（3）与地面调查结果比较,InSAR识别的滑坡更全面、规模更准确,对高位滑坡和集中分布滑坡识别更有效,可以作为现今地质灾害排查的重要手段之一;（4）雷波县地质灾害主要分布在金沙江及美姑河沿岸,北向坡和30°~40°坡度是地质灾害高发区,除寒武系、奥陶—志留系地层发生地质灾害比例较高,其它各地层总体分布较均一。      

一种基于数据的核优化新方法

针对核优化问题进行了研究,给出了一种基于数据的智能核优化新方法。算法利用UCI数据和美国实测合成孔径雷达图像数据进行仿真实验,结果验证了该方法的有效性和可行性。     

Monitoring dam structural health from space: Insights from novel InSAR techniques and multi-parametric modeling applied to the Pertusillo dam Basilicata,Italy

The availability of new constellations of synthetic aperture radar (SAR) sensors is leading to important advances in infrastructure monitoring. These constellations offer the advantage of reduced revisit times, providing low-latency data that enable analysis that can identify infrastructure instability and dynamic deformation processes.In this paper we use COSMO-SkyMed (CSK) and TerraSAR-X (TSX) data to monitor seasonal induced deformation at the Pertusillo dam (Basilicata, Italy) using multi-temporal SAR data analysis. We analyzed 198 images spanning 2010–2015 using a coherent and incoherent PS approach to merge COSMO-SkyMed adjacent tracks and TerraSAR-X acquisitions, respectively. We used hydrostatic-seasonal-temporal (HST) and hydrostatic-temperature-temporal (HTT) models to interpret the non-linear deformation at the dam wall using ground measurements together with SAR time-series analysis. Different look geometries allowed us to characterize the horizontal deformation field typically observed at dams. Within the limits of our models and the SAR acquisition sampling we found that most of the deformation at the Pertusillo dam can be explained by taking into account only thermal seasonal dilation and hydrostatic pressure. The different models show slightly different results when interpreting the aging term at the dam wall. The results highlight how short-revisit SAR satellites in combination with models widely used in the literature for interpreting pendulum and GPS data can be used for supporting structural health monitoring and provide valuable information to ground users directly involved in field measurements.     

SAR图像斑点噪声抑制方法与应用研究

通过详细分析SAR图像的判读和应用斑点噪声的算法,得出:多视处理能有效地平滑噪声,但降低空间分辨率;基于空间域滤波算法能有效地平滑斑点噪声,但不同程度地损失边缘和细节信息;基于变换域的小波多尺度滤波可以较好地保持边缘信息,然而其滤波效果不理想。在此基础上提出利用小波分析技术把两种或两种以上的单个滤波方法进行融合,在有效去除斑点噪声时能够较好地保持边缘、细节和纹理信息。最后初步讨论在实际应用中应根据不同情况和要求选择相应的滤波方法。     

Changes in surface soil mositure during freeze-thaw process under different vegetated status using Sentinel-1A in Beiluhe北大核心CSCD

Beiluhe basin lies in a permafrost region where is located in the interior of Tibetan Plateau. Ecosystem in the area is subjected to the freeze-thaw process of the active tjaele,and there is conspicuous correlation between soil moisture（SM）and vegetation coverage. To retrieve the soil moisture content of Beiluhe basin with a total area of 2 037. 94 km2,a synergistic method,which combined improved water cloud model,Oh,Dubois and Topp model,was presented in this paper base on Sentinel-1A multi-polarization SAR and Landsat-8 time series images data. The accuracy was validated with the in-situ point SM data：Adjusted-R2 of the regression equation is 0. 6848,and RMSE is 0. 039 cm3·cm-3. The analysis of correlation among freeze-thaw process,SM and vegetation cover from macro watershed scale manifests：Vegetation coverage has a significant delayed effect on the freeze-thaw process of the active tjaele,that is,the higher vegetation coverage,the more lagging freeze-thaw time;These study results are basically consistent with predecessors in-situ observation data,verifying the feasibility of studying correlation among soil freeze-thaw process,SM,and vegetation coverage from the macro watershed scale based on Sentinel-1A annual time series data. © 2022 Science Press (China).     

SAR图像河流提取的主动轮廓模型的稳健估计算法

针对现有主动轮廓模型无法精确提取SAR图像中河流的难题,提出了一种结合L₁范数和拉普拉斯能量的主动轮廓模型。首先,将Chan-Vese(CV)模型中L₂范数形式的外部能量约束项替换为L₁范数形式的外部能量约束项,得到新的能量泛函;其次,提出了一种基于拉普拉斯核函数的外部能量约束项,并将其添加到上述能量泛函中,同时赋予两种外部能量约束项不同的调节系数;最后,引入曲线内外区域像素灰度绝对中位差的均值替代模型中的常数曲线内外能量权值,以得到完整的提出模型。针对实际SAR图像进行河流提取,结果表明:与现有主动轮廓模型相比,本文提出的模型在河流提取准确性和提取效率两方面具有明显优势。     

机载X波段极化干涉SAR数据非监督分类方法研究

针对经典极化分类算法在处理机载X波段SAR数据时将过多地物分为体散射类型,并且容易受噪声影响,分类结果存在大量误分现象的问题,通过对机载X波段SAR数据非监督分类方法的研究,提出将极化干涉信息用于机载X波段极化干涉SAR数据的分类。通过运用极化干涉数据进行目标分解得到参数A1和A2对数据进行初始分类,然后结合改进的Wishart最大似然分类算法来进行地物的自适应分类。实验结果表明,该方法能有效避免平地效应的影响,抗噪性好,能正确区分三种典型散射类型,分类效果明显优于极化分类效果。     

一种利用影像模拟制作SAR立体模型的方法

针对SAR数字立体测图,提出了一种SAR理想立体模型的构建方法。利用西部横断山脉地区的机载SAR影像制作立体模型,进行数字测图试验。试验证明模型具有良好的立体视觉效果,同时能够用于高精度三维坐标的实时量测和地形图要素采集。     

Identification of potential subsidence related to pumping in the Almería basin (SE Spain)

Ground subsidence of detrital deposits in the Almería basin (SE Spain) was studied using the remote sensing technique of Differential Interferometry SAR (DInSAR). This basin is one of the most arid in Europe, receiving an average rainfall of 250 mm per year. Over the last 60 years the region has experienced an enormous agricultural and urban expansion, whose water demand has been largely supplied from groundwater, leading to the current situation of overexploitation of water resources. This paper outlines the likely relationship between groundwater abstraction and subsidence. To this end, 34 ERS and Envisat images, taken between 2003 and 2009, were analysed to estimate ground surface deformations, and hence, compared with water table variations measured in a number of piezometers in the basin. The analysis shows a clear parallelism between the variations in piezometric level and deformation of the ground surface. In addition, the zones of greatest subsidence coincide with those areas where groundwater abstractions are concentrated. Subsidence over the examined period varies from 10 to 30 mm, with extreme values as high as 50 mm, which translates to a rate of between 1·7 and 5 mm/year, reaching maximum rates of 8 mm/year at some points. Given such subsidence rates, damage to urban infrastructures are, for the moment, incipient. Copyright © 2011 John Wiley & Sons, Ltd.     

西南天山明尧勒背斜河流阶地沉积物的光释光测年

晚第四纪细颗粒河流阶地沉积物的准确定年是当前光释光测年研究中的一个难点。利用简单多片再生法(SMAR)、单片再生法(SAR)等对新疆西南天山喀什前陆盆地明尧勒背斜喀浪勾律克河阶地沉积物样品的细颗粒石英光释光(BLSL)信号、混合矿物红外释光信号(IRSL)和红外后蓝光释光信号([Post-IR]OSL)进行测量。通过对比同一样品的细颗粒石英BLSL、混合矿物IRSL和[Post-IR]OSL的等效剂量值,认为细颗粒石英简单多片再生法是河流阶地沉积物释光测年最为适用的方法之一。对于细颗粒混合矿物而言,由于长石矿物的异常衰退,对于等效剂量De>70Gy的样品,其IRSL等效剂量值均偏小;其[Post-IR]OSL等效剂量值仍需慎重使用。通过对同一样品细颗粒石英多个测片的单片再生法等效剂量值分布以及简单多片再生法多个测片校正后天然释光信号强度的一致性判断,认为样品LEDL03-87q,LEDL03-86q和LEDL04-54q在沉积时的光晒退是不均匀的,采用最小等效剂量组分来计算沉积年龄