Satellite SAR Images Research Articles

Analysis of tectonic activity of the North Anatolian Fault based on SBInSAR method

The North Anatolian Fault situated in Turkey is one of the longest and most active tectonic faults in the world. The paper presents an analysis of tectonic activity in its area based on the method of Small Baseline Synthetic Aperture Radar Interferometry. For this purpose 73 satellite SAR images and specialized software GMT5SAR were used with implement the SBAS algorithm. In addition, the most important aspects of data processing and their final products were presented, which determined the surface displacements occurring in the surveyed area from 1 January 2014 to 1 March 2017. The displacements of the SBAS surface area ranged from -10 cm to +10 cm. Based on the obtained results and their analysis, the author also assessed the suitability of SBInSAR technology for areas of land displacement.

Use of remote sensing data in comprehending an extremely unusual flooding event over southwest Bangladesh

Flooding is one of the natural disasters that affect the livelihood of the people living in the floodplains, like Bangladesh. Here, we proposed to employ SAR satellite images in assessing the flood extent and crop damage with the hydrometeorological observations in the southwestern region of Bangladesh. We observed that the unusual flood of the year 2000 was the combined effect of the huge amount of rainfall in the local areas as well as oncoming water flows from West Bengal in India. During late monsoon of 2000, we experienced that the amount of rainfall was in several magnitudes (250–450%) than the expected over the region. Bangladesh, one of the largest deltas in the world, in general experienced recurrent flood events from the spill of the three mighty rivers in every year. However, we observed that during 2000 the river situation was slightly above the average year conditions and below the moderate level of flood warning conditions, indicating non-riverine flooding. Therefore, we used the SAR images in delineating the flood extent and its damages for the standing aman crops. We observed that the flood extent mapping was having more than 95% agreements with the ground data and crop damage information was about 75% in agreement with the government estimates. The flood extent and crop damage map was found to be useful during the unprecedented flood in the southwestern region of Bangladesh. The use of near real-time SAR imageries thus would be helpful in developing strategies for flood management and disaster mitigation activities and could be utilized on a regular basis.

FOSS4G DATE ASSESSMENT ON THE ISPRS OPTICAL STEREO SATELLITE DATA: A BENCHMARK FOR DSM GENERATION

Abstract. The ISPRS Working Group 4 Commission I on “Geometric and Radiometric Modelling of Optical Spaceborne Sensors”, provides a benchmark dataset with several stereo data sets from space borne stereo sensors. In this work, the Worldview-1 and Cartosat-1 datasets are used, in order to test the Free and Open Source Software for Geospatial (FOSS4G) Digital Automatic Terrain Extractor (DATE), developed at Geodesy and Geomatics Division, University of Rome “La Sapienza”, able to generate Digital Surface Models starting from optical and SAR satellite images. The accuracy in terms of NMAD ranges from 1 to 3 m for Wordview-1, and from 4 to 6 m for Cartosat-1. The results obtained show a general better 3D reconstruction for Worldview-1 DSMs with respect to Cartosat-1, and a different completeness level for the three analysed tiles, characterized by different slopes and land cover.

P54 火山噴火時の情報収集におけるSAR衛星画像の適用について(ポスターセッション,日本火山学会2008年秋季大会)

P54 火山噴火時の情報収集におけるSAR衛星画像の適用について(ポスターセッション,日本火山学会2008年秋季大会)

Combined observations of Arctic sea ice with near‐coincident colocated X‐band, C‐band, and L‐band SAR satellite remote sensing and helicopter‐borne measurements

AbstractIn this study, we compare colocated near‐coincident X‐, C‐, and L‐band fully polarimetry SAR satellite images with helicopter‐borne ice thickness measurements acquired during the Norwegian Young sea ICE 2015 (N‐ICE2015) expedition in the region of the Arctic Ocean north of Svalbard in April 2015. The air‐borne surveys provide near‐coincident snow plus ice thickness, surface roughness data, and photographs. This unique data set allows us to investigate how the different frequencies can complement one another for sea ice studies, but also to raise awareness of limitations. X‐band and L‐band satellite scenes were shown to be a useful complement to the standard SAR frequency for sea ice monitoring (C‐band) for lead ice and newly formed sea ice identification. This may be in part be due to the frequency but also the high spatial resolution of these sensors. We found a relatively low correlation between snow plus ice thickness and surface roughness. Therefore, in our dataset ice thickness cannot directly be observed by SAR which has important implications for operational ice charting based on automatic segmentation.

Best Accuracy Land Use/Land Cover (LULC) Classification to Derive Crop Types Using Multitemporal, Multisensor, and Multi-Polarization SAR Satellite Images

When using microwave remote sensing for land use/land cover (LULC) classifications, there are a wide variety of imaging parameters to choose from, such as wavelength, imaging mode, incidence angle, spatial resolution, and coverage. There is still a need for further study of the combination, comparison, and quantification of the potential of multiple diverse radar images for LULC classifications. Our study site, the Qixing farm in Heilongjiang province, China, is especially suitable to demonstrate this. As in most rice growing regions, there is a high cloud cover during the growing season, making LULC from optical images unreliable. From the study year 2009, we obtained nine TerraSAR-X, two Radarsat-2, one Envisat-ASAR, and an optical FORMOSAT-2 image, which is mainly used for comparison, but also for a combination. To evaluate the potential of the input images and derive LULC with the highest possible precision, two classifiers were used: the well-established Maximum Likelihood classifier, which was optimized to find those input bands, yielding the highest precision, and the random forest classifier. The resulting highly accurate LULC-maps for the whole farm with a spatial resolution as high as 8 m demonstrate the beneficial use of a combination of x- and c-band microwave data, the potential of multitemporal very high resolution multi-polarization TerraSAR-X data, and the profitable integration and comparison of microwave and optical remote sensing images for LULC classifications.

Remote Sensing Image Classification: No Features, No Clustering

In this paper, we consider the problem of remote sensing image classification, in which feature extraction and feature coding are critical steps. Various feature extraction methods aim at an abstract and discriminative image representation. Most of them are either theoretically too complex or practically infeasible to compute for large datasets. Motivated by this observation, we propose a simple yet efficient feature extraction method within the bag-of-words (BoW) framework. It has two main innovations. First and most interestingly, this method does not need any complex local feature extraction; instead, it uses directly the pixel values from a local window as low level features. Second, in contrast to many unsupervised feature learning methods, a random dictionary is applied to feature space quantization. The advantage of a random dictionary is that it does not need the time-consuming process of dictionary learning yet without a significant loss of classification accuracy. These two novel improvements over state-of-the-art methods significantly reduce the computational time and enable it scalable to a large data volume. An extensive experimental evaluation has been performed and compared with other feature extraction methods. It is demonstrated that our feature extraction method is quite competitive and can achieve rather promising performance figures for both optical and SAR satellite images.

An integrated method based on DInSAR, MAI and displacement gradient tensor for mapping the 3D coseismic deformation field related to the 2011 Tarlay earthquake (Myanmar)

Abstract The satellite differential interferometric synthetic aperture radar (DInSAR) technology has been widely applied for mapping the ground deformation associated with the geophysical events such as earthquakes. However, the conventional DInSAR can measure only the one-dimensional (1D) ground displacement along the radar line of sight (LOS), and the crucial displacement measurements, e.g., near the epicenter and along the surface ruptures, are usually not available or degraded in quality due to the significant deformation gradients. With availability of satellite SAR images acquired along ascending and descending orbits, this paper proposes an integrated method to map the three-dimensional (3D) coseismic deformation field by combining DInSAR for detecting LOS displacements, the multiple aperture interferometry (MAI) for detecting along-track displacements, and the displacement gradient tensor (DGT) model for characterizing spatial correlation of ground displacements. The proposed method (termed as InSAR–DGT) was first tested through an experiment of recovering 3D displacements from a simulated coseismic deformation field. We then applied the proposed method to map the 3D coseismic deformation field related to the 2011 Tarlay earthquake (Myanmar) by using the ascending and descending ALOS PALSAR images. Both the results derived for the simulated experiment and the 2011 Tarlay earthquake show that the quality of the 3D coseismic displacements can be raised efficiently by the InSAR–DGT method, and the precisions in the east–west (E–W), north–south (N–S) and up–down (U–D) displacements for the 2011 Tarlay earthquake are increased 22%, 36% and 24%, respectively. The validation indicates that the improved coseismic deformation field for the 2011 Tarlay earthquake is in good agreement with the deformation field simulated from the existing optimized fault model. The number of the missing data points in the 3D coseismic deformation field can be reduced significantly by the InSAR–DGT method. It is revealed that the causative fault (i.e., the west part of the Nam Ma Fault) for the Tarlay earthquake generated a left-lateral slip that was accompanied by a minor normal dip-slip component. The careful interpretation demonstrates that the causative fault structures include the determined main fault segment and the two suspected branched segments at east of the Tarlay town.

Modified four-pass differential SAR interferometry for estimating mountain glacier surface velocity fields

ABSTRACTConventional four-pass differential synthetic aperture radar interferometry (DInSAR) assumes that there are no significant changes in the ground during the period between the acquisition times of SAR images for topographic DInSAR pairs. This assumption can rarely be satisfied for glacial areas due to their continuous movement. This letter proposes a modified four-pass DInSAR method without an external digital elevation model (DEM), taking into account glacier movement between the acquisition times of SAR images used to form topographic DInSAR pairs. An explicit expression of theoretical formulas for a modified four-pass DInSAR technique was derived for the first time, revealing that four-pass DInSAR considering ground movement of topographic pairs was equivalent to that of conventional four-pass DInSAR with a spatially varying nominal wavelength. Then the proposed method was tested with four Advanced Land Observing Satellite (ALOS) SAR images covering Dongkemadi glacier located on the Tibetan Plateau, China. An experiment with real data showed that the proposed method could obtain glacial flow patterns efficiently, and that the difference between two-pass DInSAR and the proposed method is a result of DEM bias and glacial thinning. The approach presented in this letter proved to be appropriate for monitoring glacial motion and provides a valuable tool for glacier studies, without the need of an external DEM.

Understanding the subsidence process of a quaternary plain by combining geological and hydrogeological modelling with satellite InSAR data: The Acque Albule Plain case study

Abstract This paper focuses on a multidisciplinary study carried out in an urban area affected by subsidence and related structural damages. The study area is located about 20 km east of Rome (Italy) and is characterised by relevant groundwater exploitation for various purposes as well as by the presence of compressible soils immediately below the ground level. Extensive processing at different scales of SAR satellite images (ERS and ENVISAT provided by ESA in the frame of a CAT-1 project) by means of A-DInSAR technique was performed. The time histories of ground displacements, have been analysed in combination with a detailed geological setting of the study area and with the hydrogeological changes occurred in the last decades (as the response to the anthropic stress) based on a large piezometric dataset. This comprehensive dataset allowed us to describe the space and time distributions of the subsidence process. The spatial pattern and deformation rate change is attributed to the following causes: i) the changes in the groundwater levels due to the intensification of mine exploitation (requiring dewatering operations) and ii) the distribution and thickness of recent compressible deposits. Specifically, it is derived that the groundwater level variations drive the timing of subsidence triggering over the area, whereas the local geological conditions control the magnitude of the deformation process.

Land subsidence, Ground Fissures and Buried Faults: InSAR Monitoring of Ciudad Guzmán (Jalisco, Mexico)

We study land subsidence processes and the associated ground fissuring, affecting an active graben filled by thick unconsolidated deposits by means of InSAR techniques and fieldwork. On 21 September 2012, Ciudad Guzmán (Jalisco, Mexico) was struck by ground fissures of about 1.5 km of length, causing the deformation of the roads and the propagation of fissures in adjacent buildings. The field survey showed that fissures alignment is coincident with the escarpments produced on 19 September 1985, when a strong earthquake with magnitude 8.1 struck central Mexico. In order to detect and map the spatio-temporal features of the processes that led to the 2012 ground fissures, we applied InSAR multi-temporal techniques to process ENVISAT-ASAR and RADARSAT-2 satellite SAR images acquired between 2003 and 2012. We detect up to 20 mm/year of subsidence of the northwestern part of Ciudad Guzmán. These incremental movements are consistent with the ground fissures observed in 2012. Based on interferometric results, field data and 2D numerical model, we suggest that ground deformations and fissuring are due to the presence of areal subsidence correlated with variable sediment thickness and differential compaction, partly driven by the exploitation of the aquifers and controlled by the distribution and position of buried faults.

고해상도 SAR 영상 및 EO 영상을 이용한 표적군 검출 기법 개발

본 연구에서는 고해상도 위성영상인 TerraSAR-X와 WorldView-2 등을 융합하여 표적의 특성을 고려한 표적군(Group of targets) 검출을 수행하였다. 관심 대상으로 하는 표적은 고정되어 있으며, 군(Group)을 이루고 있는 특징이 있다. 표적 후보를 검출하기 위해 대상 물체의 레이더 후방산란 특성을 이용한 Constant False Alarm Rate (CFAR) 알고리즘을 적용하였다. 검출된 표적 후보군으로부터 비표적을 제거하기 위해 표적의 크기 정보를 이용한 화소 클러스터링, 표적군을 이루는 표적들간의 배치 특성을 이용한 네트워크 클러스터링. 그리고 SAR 간섭기법 적용이 가능한 간섭쌍이 있는 경우 긴밀도 정보를 이용하였다. 또한, 오경보(False Alarm)를 감소시키고 최종 표적을 결정하기 위해, 표적의 형태 정보를 추출할 수 있는 Electro-Optical (EO) 영상을 바탕으로 효과적인 타원 검출 기법을 개발하였다. 개발된 표적군 검출 알고리즘을 10개 지역에 적용한 결과, 표적군 검출율은 100%, 단일 표적에 대한 오경보율은 0.03~0.3개/<TEX>$km^2$</TEX>, 평균 오경보는 1.8군/<TEX>$64km^2$</TEX>로 낮은 오경보와 높은 검출율을 보이며 표적군이 검출되었다. 본 연구에서 개발된 표준화된 표적 검출 기법은 향후 무인화된 표적 검출 시스템 구축에 핵심적인 기술이 될 것으로 전망한다. In this study, the target detection using both high-resolution satellite SAR and Elecro-Optical (EO) images such as TerraSAR-X and WorldView-2 is performed, considering the characteristics of targets. The targets of our interest are featured by being stationary and appearing as cluster targets. After the target detection of SAR image by using Constant False Alarm Rate (CFAR) algorithm, a series of processes is performed in order to reduce false alarms, including pixel clustering, network clustering and coherence analysis. We extend further our algorithm by adopting the fast and effective ellipse detection in EO image using randomized hough transform, which is significantly reducing the number of false alarms. The performance of proposed algorithm has been tested and analyzed on TerraSAR-X SAR and WordView-2 EO images. As a result, the average false alarm for group of targets is 1.8 groups/<TEX>$64km^2$</TEX> and the false alarms of single target range from 0.03 to 0.3 targets/<TEX>$km^2$</TEX>. The results show that groups of targets are successfully identified with very low false alarms.

Oil Spill Monitoring Based on SAR Remote Sensing Imagery

Compared with traditional on-site oil spill monitoring, the use of remote sensing technology has the macroscopic characteristics, which could quickly and accurately find an oil spill area. Currently, the detection approach of monitoring the phenomenon of marine oil spill are usually divided into two types, which are optical and synthetic aperture radar SAR remote sensing imagery. Among all satellite sensors, SAR is still the most utilized for operational oil spill detection. Thus, adopt SAR imagery to achieve routinely monitoring. In the image analysis process, the discrimination of oil spills and look-alike phenomena e.g., low wind area, wind front area and natural slicks on SAR is a crucial task in marine oil spill. A support vector machine is employed to remote sensing image classification in this paper. Through the simulation of the Dalian oil spill event, the effectiveness of the proposed approach for SAR satellite image classification is verified.

X-band SAR 위성영상의 3차원 위치결정 기법 평가 - 레이더 간섭기법, Radargrammetry, RPC를 중심으로

Korea`s first X-band SAR satellite KOMPSAT-5 has been launched in 2013, so the research related to the X-band SAR satellite image is required to increase the utilization of KOMPSAT-5. In this study, we generated a DEM(Digital Elevation Model) using X-band SAR satellite images based on three methods which are InSAR, radargrammetry and RPC(Rational Polynomial Coefficients), and evaluated the performance of each methods. The four stripmap mode TerraSAR-X images taken in Daejeon were used to generate DEM, and accuracy was evaluated using DEM by IKONOS RPC. As results, DEM produced by the InSAR showed the highest accuracy. Also, we knew that RPC could be effective method if you want to create a large area DEM which contains the various elevation.

A Hierarchical Approach to Persistent Scatterer Network Construction and Deformation Time Series Estimation

This paper presents a hierarchical approach to network construction and time series estimation in persistent scatterer interferometry (PSI) for deformation analysis using the time series of high-resolution satellite SAR images. To balance between computational efficiency and solution accuracy, a dividing and conquering algorithm (i.e., two levels of PS networking and solution) is proposed for extracting deformation rates of a study area. The algorithm has been tested using 40 high-resolution TerraSAR-X images collected between 2009 and 2010 over Tianjin in China for subsidence analysis, and validated by using the ground-based leveling measurements. The experimental results indicate that the hierarchical approach can remarkably reduce computing time and memory requirements, and the subsidence measurements derived from the hierarchical solution are in good agreement with the leveling data.

Rapid Disaster Analysis based on Remote Sensing: A Case Study about the Tohoku Tsunami Disaster 2011

Abstract. In this study, we present first results of RAPIDMAP, a project funded by European Union in a framework aiming to foster the cooperation of European countries with Japan in R&amp;amp;D. The main objective of RAPIDMAP is to construct a Decision Support System (DSS) based on remote sensing data and WebGIS technologies, where users can easily access real-time information assisting with disaster analysis. In this paper, we present a case study of the Tohoku Tsunami Disaster 2011. We address two approaches namely change detection based on SAR data and co-registration of optical and SAR satellite images. With respect to SAR data, our efforts are subdivided into three parts: (1) initial coarse change detection for entire area, (2) flood area detection, and (3) linearfeature change detection. The investigations are based on pre- and post-event TerraSAR-X images. In (1), two pre- and post-event TerraSAR-X images are accurately co-registered and radiometrically calibrated. Data are fused in a false-color image that provides a quick and rough overview of potential changes, which is useful for initial decision making and identifying areas worthwhile to be analysed further in more depth. However, a bunch of inevitable false alarms appear within the scene caused by speckle, temporal decorrelation, co-registration inaccuracy and so on. In (2), the post-event TerraSAR-X data are used to extract the flood area by using thresholding and morphological approaches. The validated result indicates that using SAR data combining with suitable morphological approaches is a quick and effective way to detect flood area. Except for usage of SAR data, the false-color image composed of optical images are also used to detect flood area for further exploration in this part. In (3), Curvelet filtering is applied in the difference image of pre- and post-event TerraSAR-X images not only to suppress false alarms of irregular-features, but also to enhance the change signals of linear-features (e.g. buildings) in settlements. Afterwards, thresholding is exploited to extract the linear-feature changes. In rapid mapping of disasters various sensors are often employed, including optical and SAR, since they provide complementary information. Such data needs to be analyzed in an integrated fashion and the results from each dataset should be integrated in a GIS with a common coordinate reference system. Thus, if no orthoimages can be generated, the images should be co-registered employing matching of common features. We present results of co-registration between optical (FORMOSAT-2) and TerraSAR-X images based on different matching methods, and also techniques for detecting and eliminating matching errors.

Transformation of internal solitary waves at the &amp;quot;deep&amp;quot; and &amp;quot;shallow&amp;quot; shelf: satellite observations and laboratory experiment

Abstract. An interaction of internal solitary waves with the shelf edge in the time periods related to the presence of a pronounced seasonal pycnocline in the Red Sea and in the Alboran Sea is analysed via satellite photos and SAR images. Laboratory data on transformation of a solitary wave of depression while passing along the transverse bottom step were obtained in a tank with a two-layer stratified fluid. The certain difference between two characteristic types of hydrophysical phenomena was revealed both in the field observations and in experiments. The hydrological conditions for these two processes were named the "deep" and the "shallow" shelf respectively. The first one provides the generation of the secondary periodic short internal waves – "runaway" edge waves – due to change in the polarity of a part of a soliton approaching the shelf normally. Another one causes a periodic shear flow in the upper quasi-homogeneous water layer with the period of incident solitary wave. The strength of the revealed mechanisms depends on the thickness of the water layer between the pycnocline and the shelf bottom as well as on the amplitude of the incident solitary wave.

Evaluation of SAR wind retrieval algorithms in offshore areas of the Korean Peninsula

The method of extracting offshore wind field from SAR satellite images for offshore wind resource assessment shows some discrepancies in the calculated wind speed depending on which of the CMOD algorithms is used. This paper compared different algorithms, such as CMOD_IFR2, CMOD4, CMOD5, in order to find the suitable CMOD algorithm for the southern offshore area of the Korean Peninsula, which belongs to the zone influenced by the archipelago, by calculating the relative errors with reference to the Korea wind map of representative weather days by season. The result is that CMOD-IFR2 shows a tendency towards excessive estimation of wind speed, while CMOD4 shows the lowest RMSE values of wind speed differences with respect to the Korea wind map. Based on this result, we have concluded that, for the southern offshore area of the Korean Peninsula which belongs to the medium-range wind speed zone, CMOD4 is the suitable algorithm.

Segmentation of SAR Satellite Images Using Cellular Learning Automata and Adaptive Chains

This paper presents a new method to segment SAR satellite images that depends on hard and easy segmentation processes. The proposed algorithm compounds the intensity information and input image texture using Cellular Learning Automata (CLA) in order to segment SAR images. In many of the segmentation techniques, some standards are first defined, and then the segmentation process acts according to this standards and neighbor image information. Information can be easily simulated and performed by CLA. Our proposed method performs in two steps: easy and hard segmentation. In easy segmentation step, input SAR satellite image is analyzed by CLA and we try to decrease intensity differences between pixels in same repeated region and produce an easy segmentation image, then this image is given to a hard segmentation system. Obtained results show that the proposed method is useful. KeywordsSynthetic Aperture Radar; SAR Image; Learning Automata; Cellular Learning Automata

高解像度衛星SAR 画像を用いた建物側面の被害把握

高解像度衛星SAR 画像を用いた建物側面の被害把握