基于多方向多源合成孔径雷达数据融合的假彩色正射影像制作

在复杂地形条件下,SAR影像几何失真大,辐射畸变突出,透视收缩、叠掩和阴影等现象严重,制约SAR测图的应用.为此,提出一套基于多方向多源SAR数据融合方法与技术流程,利用不同侧视方向、不同分辨率、多极化的SAR影像数据,融合制作信息量丰富,能够满足应用要求的SAR假彩色正射影像.采用横断山脉区域3 m分辨率的Terra...     

航空重力测量在近海区域的精度评估与分析

利用泊松积分法和点质量法对澳大利亚West Arnhem Land区域的航空重力测量数据进行了精度评估,两种方法得到精度结果基本一致,评估结果表明GT-1A测量系统2′分辨率数据的测量精度优于3×10-5 m/s2,5′分辨率数据的测量精度优于2×10-5 m/s2。利用交叉点平差和泊松积分法、点质量法对渤海区域的航空重力测量进行了内部交叉点平差和外部精度评估,结果表明,内部评估精度与外部评估精度存在一定的差异,以外部评估为准则,CHAGS测量系统在渤海区域5′分辨率的航空重力数据精度优于3.5×10-5 m/s2。综合国内外试验情况分析得到,在近海区域,航空重力数据的分辨率和精度受测量仪器的性能而不同,整体上对于5′分辨率数据而言,可以达到或优于3×10-5 m/s2的精度。     

波罗的海高精度海洋重力测量数据处理与分析

海洋重力测量是海洋地区获取高精度、高分辨率地球重力场信息的重要技术方法。研究了波罗的海地区海洋重力测量数据的处理方法,利用测线交叉点处重力测量观测值建立的漂移函数有效减小了重力仪漂移异常的影响,其测量精度达到了0.5 mGal。2017年,海空重力仪Chekan-AM更新以后,其测量稳定性有了显著提高。数据处理结果表明,搭载在常规科考船DENEB的海洋重力测量精度为0.2 mGal,而首次搭载在渡轮URD的海洋重力测量精度为0.6 mGal。根据波罗的海地区局部大地水准面构建的初步结果,证明了获取的海洋重力测量数据对填补数据空白、检测老旧数据以及提高大地水准面精度等方面具有重要作用,研究结果为未来波罗的海地区建立高精度的统一大地水准面提供了数据基础和技术支持。     

天宫二号近天底角交轨干涉SAR的海洋涡旋探测

海洋涡旋是海洋科学研究的一个重要分支，在海洋热循环中起着重要的作用。传统的卫星高度计通常能够探测到水平尺度超过100 km的中尺度涡旋，但是由于其分辨率较为粗糙，对于1—100 km的亚中、小尺度涡旋难以探测。天宫二号搭载的三维成像微波高度计（InIRA）是国际上第1个可以用于进行海面高度测量的Ku波段干涉SAR，为观测和研究海洋亚中、小尺度涡旋提供了数据来源。InIRA不仅可以获取海洋涡旋2维SAR图像，同时通过两个天线获取复图像间的干涉相位，能够获取涡旋海表面高度异常，为海洋涡旋的探测研究提供了新的可能。本文基于天宫二号InIRA数据，开展了海洋涡旋探测的研究，提出一种通过计算涡旋海面相对高度变化实现海洋涡旋探测的方法。通过对InIRA涡旋数据的处理与分析，发现该方法能够实现海洋涡旋的探测，并利用MODIS叶绿素浓度数据和海表温度数据对海洋涡旋探测结果进行了验证。     

一种结合上下文与边缘注意力的SAR图像海陆分割深度网络方法

海陆分割对于合成孔径雷达(synthetic aperture radar,SAR)图像海洋目标检测、海岸线提取等任务具有重要意义。针对实际应用中多分辨率SAR图像海陆分割难题,提出了一种基于上下文与边缘注意力的海陆分割方法。该方法利用通道注意力机制融合不同尺度和层次的上下文特征,设计了边缘提取支路提供边缘信息,进一步提高了海陆边界的分割准确率。同时,构建了基于高分三号卫星数据的多分辨率SAR图像海陆分割数据集,该数据集涵盖了多个分辨率,包括港口、岛屿等多种海陆边界类型。并基于所构建的多分辨率SAR图像海陆分割数据集,对所提网络的有效性和各模块的作用进行了实验分析。实验结果表明,所提网络的整体预测准确率和平均交并比分别达到了98.21%和96.47%,能够较好地完成海陆分割任务。     

基于Faster R-CNN的卫星SAR图像南海海洋内波自动检测

海洋内波广泛存在于世界各大洋和边缘海中,在海洋能量串级中扮演着重要角色,在海洋资源开发、海洋工程建设和海洋军事活动等方面均具有重要学术价值与实际意义。海洋内波在合成孔径雷达SAR (Synthetic Aperture Radar)图像上呈现出亮暗相间的条纹状特征。本文利用2001年—2020年南海海域包含不同微波波段(C、L、X)、不同极化方式、不同空间分辨率的631幅星载SAR图像,构建了5480个SAR图像南海海洋内波样本,结合Faster R-CNN框架,利用迁移学习的方法,实现SAR图像上的海洋内波自动检测。模型识别准确率达到95.7%,召回率为92.3%,在准确率较高的同时还能保持较低的虚警率。该算法的建立使得基于海量卫星SAR数据检出海洋内波成为可能,从而为针对性地开展内波动力参数反演和过程研究提供了技术和数据基础。     

应用雷达干涉技术的交通设施形变监测研究

针对常规测量手段耗时、费力,难以快速完成大型交通设施的形变监测的问题,该文开展了基于高空间分辨率SAR影像的PS-InSAR在大型交通设施形变监测方面的实验研究:选取2009—2010年间27景3m分辨率的TerraSAR-X数据,采用PS-InSAR技术提取了天津市西部城郊结合区内主要交通设施及沿线区域的地表形变信息。结果表明:利用高分辨率SAR影像监测公路、铁路等交通设施时,可以提取高密度的PS监测点;提取形变信息与水准观测值相互吻合;研究区域存在几个较大的沉降中心,使得一些交通设施在部分路段沉降较大。     

改进的ELU卷积神经网络在SAR图像舰船检测中的应用

随着航天技术的发展,我国SAR载荷的探测体系呈现多种类、多分辨率的发展趋势。传统的检测识别方法很难适应多分辨率、多种类的SAR图像数据,从而需要寻求一种能从多分辨率的图像数据中提取有效特征的方法。智能化发展非常迅速,本文基于SAR图像的特点,提出了改进的ELU激活函数卷积神经网络的方法,建立了结合ELU激活函数和二次代价函数的深度学习模型。同时,在训练样本中建立样本特征与所在分类中心的距离函数,用模糊支持向量机（FSVM）对提取的特征进行了分类。试验结果表明,本文方法提高了SAR图像舰船检测的抗噪性,并且检测率达到了98.6%。     

利用全极化SAR数据去方位向模糊的舰船目标检测方法

在SAR成像过程中,利用目标相对于雷达运动的多普勒频率进行方位向压缩来提高方位向分辨率。对于海面舰船目标,往往由于SAR发射的脉冲重复频率过低,回波信号的多普勒频率欠采样,引起高亮度的方位向模糊噪声,产生虚假目标。若直接利用传统CFAR、子孔径相干等目标检测算法,会把方位向模糊引起的虚假目标误判为舰船目标,降低了舰船目标的检测精度。针对SAR影像方位向模糊现象,提出一种利用全极化SAR数据的舰船目标检测算法,其利用H-A-Alpha分解得到的第三散射机制,结合方位向全极化子孔径相干,消除了方位向模糊噪声对舰船目标检测的影响,提高了舰船目标的目标杂波比。通过采用日本玉野Kojimawan港口附近的C波段AIRSAR全极化数据进行试验,验证了该算法能消除SAR方位向模糊,降低虚警率,提高舰船目标的检测精度。     

轻便型移动测量系统在大比例尺DOM质检中的应用

为研究车载移动测量技术在大比例尺DOM质量检验中的应用,本文以一种轻便型车载移动测量系统为例,首先介绍了系统及其误差来源,并通过已有数据检测分析系统的测量精度,继而利用该系统对城区0.2 m分辨率试验DOM进行点云采集、检测点提取和精度评定,结果表明利用轻便型移动测量系统进行大比例尺DOM的质量检验是可行的。     

基于SAR图像的海洋溢油分割方法研究

目前,海洋溢油污染在发生频率、分布范围以及所带来的危害程度上均居海洋污染首位。它对沿海城市生态、环境、经济的可持续发展造成严重破坏,因此越来越受到全社会的广泛关注。从众多海洋溢油分割方法中选取边缘检测算子和种子填充算法对SAR图像进行分割。首先运用欧空局的BEST软件进行数据转换以及图像预处理,然后使用ENVI软件对图像进行滤波处理。再用ERDAS软件进行灰度均衡和分割算法的实施,并针对算法的一些缺陷,进一步加入纹理分析使图像中的油膜与类油膜以及海水与海浪等更好的区分开。最后对不同的图像效果进行总结和比较。     

SAR影像中海洋浮油膜特征分割的Level Set方法

介绍了一种全新的区域影像分割技术--基于迎风格式偏微分方程（PDE）的Level Set方法进行海洋浮油膜特征提取.在该方法中,海洋浮油膜特征表示为扩散界面,影像灰度的梯度决定了界面扩散的方向和强度.界面边缘在影像灰度差异动力和曲率流的共同作用下不断向外扩散,能够有效地克服尖锐突起和裂缝等,在特征边缘趋于稳定和光滑.同时,该方法对高噪声具有一定抑制作用,适用于低对比度高噪声图像,特别是SAR图象中特征提取.以不同地区ERS-2 SAR图像中海洋浮油膜提取为例,进行方法验证,同时对传统的影像分割技术进行对比.     

Detection and monitoring of offshore oil seeps using ERS/ENVISAT SAR/ASAR data and seep-seismic studies in Krishna–Godavari offshore basin,India

An attempt has been made to use synthetic aperture radar (SAR) data for detection and monitoring of offshore oil seeps in the eastern offshore areas of the Krishna–Godavari Basin, which has been supplemented and correlated with collateral free-air gravity and seismic data. Images of the study area obtained from ENVISAT ASAR image mode were processed and analysed in detail. A number of natural oil seepages were identified and distinguished from pollution and biogenic slicks. These were subsequently studied using different parameters to assign various degrees of confidence. The repetitiveness of the identified seepages was studied and a total of five areas of seep repetitions had been recognized in the study area. The seeps that are repeated in images of different dates are more likely to be of natural origin than others. Simulation and modelling of a particular oil slick arising has been attempted over the Krishna–Godavari offshore using MIKE 21 software.     

On the degree of polarization for SAR sea oil slick observation

A polarimetric model to relate the degree of polarization, DoP, to the sea surface scattering with and without oil slicks, under low-to-moderate wind conditions, is proposed. DoP, measured directly from the Mueller scattering matrix, is shown to be a reliable measure of the departure from Bragg scattering; a phenomenon that, under low-to-moderate wind conditions, occurs when an oil slick is present. Following this theoretical rationale, a simple filter is developed to observe oil slicks in quad-polarimetric full-resolution Synthetic Aperture Radar (SAR) data. Experiments, undertaken on a meaningful set of quad-polarization Single Look Complex (SLC) C-band RADARSAT-2 SAR data, where both well-known oil slicks and a weak-damping look-alike are in place, demonstrate the soundness of the model and its effectiveness from an operational viewpoint.     

Determining oil slick thickness using hyperspectral remote sensing in the Bohai Sea of China

Abstract

Determining oil slick thickness plays an important role in assessing oil spill volume and its environmental impacts on the ocean. In this study, we used a Hyperion image of an oil spill accident area and seawater and fresh crude oil samples collected in the Bohai Sea of China. A well-controlled laboratory experiment was designed to simulate spectral responses to different oil slick thicknesses. Spectral resampling and normalization methods were used to reduce the differences in spectral reflectances between the experimental background seawater sample and real background seawater. Fitting the analysis with laboratory experimental data results showed a linear relationship between normalized oil slick reflectance and normalized oil slick thickness [20th band (R ²=0.92938, n=49, p<0.01), 26th band (R ²=0.93806, n=49, p<0.01), 29th band (R ²=0.93288, n=49, p<0.01)]. By using these statistical models, we successfully determined the normalized oil slick thickness with the Hyperion image. Our results indicate that hyperspectral remote sensing technology is an effective method to monitor oil spills on water. The spectral ranges of visible green and red light were the optimal bands for estimating oil slick thickness in case 2 water. The high, stabilized spectral reflectance of background seawater will be helpful in oil slick thickness inversion.     

Automatic Detection of Oil Spill Disasters Along Gulf of Mexico Using RADARSAT-2 SAR Data

In this work, the genetic algorithm is utilised for automatic detection of oil spills under wind speed conditions larger than 20 ms^?1. The procedure is implemented using sequences of RADARSAT-2 SAR ScanSAR Narrow single beam data acquired in the Gulf of Mexico. The study demonstrates that implementing crossover allows for generation of accurate oil spills pattern. This conclusion is confirmed by the receiver–operating characteristic (ROC) curve. The ROC curve indicates that the existence of oil slick footprints can be identified with the area under the ROC curve and the no-discrimination line of 85 %, which is greater than that of other surrounding environmental features. In conclusion, the genetic algorithm can be used as a tool for the automatic detection of oil spills and the ScanSAR Narrow single beam mode serves as an excellent sensor for oil spill detection and surveying under wind speed larger than 20 ms^?1.     

Oil slick identification from imageries using optical power spectra

A new method for semi-automatic identification of oil slick on ocean/sea surface from space borne imagery using optical power spectra has been suggested. This method is based on the fact that the presence of the oil slick in the ocean suppresses the normal wave pattern that will be present in the ocean area without oil slick and the power spectra of the image area having the oil slick will be devoid of the peak corresponding to the dominant wave. In addition, the general shape of the oil slick will increase the low frequency content in the power spectra. The method has been applied for test areas containing oil slicks and the results are encouraging.     

Study of dems derived from ERS-1/2 SAR and SRTM data

The DEM of the Bhuj earthquake affected area of 50 x 50 km was generated using the ERS-1/2 SAR tandem data (May 15—16,1996). Region growing algorithm coupled with path following approach was used for phase unwrapping. Phase to height conversion was done using D-GPS control points. Geocoding was done using GAMMA software. A sample data of DEM of Shuttle Radar Topography Mission (SRTM) of the Bhuj area is made available by DLR Germany. The intensity image, DEM and Error map are well registered. The spatial resolution of this DEM is about 25 m with height accuracy of a few meters. The DEM derived through ERS SAR data is prone to atmospheric affects as the required two images are acquired in different timings where as SRTM acquired the two images simultaneously. An RMS height error of 12.06 m is observed with reference to SRTM though some of the individual locations differ by as much as 35 m.     

SAR图像不同油膜特征的最佳分割方法选择

SAR图像溢油分割是SAR溢油监测中一个重要环节。文中选取4种不同形状、尺寸和对比度的SAR油膜数据,分别采用双峰阈值分割法、最大熵分割法、区域生长法、分水岭算法、图割法、水平集方法等6种方法进行溢油信息提取,探讨适合于不同油膜特征的最佳提取方法。结合现有尺度分割标准,提出一种SAR图像溢油信息评价指数——有效分割指数(Effective Segmentation Index,ESI),对不同分割方法得到的溢油提取结果进行定量评价,得出了不同特征油膜所适合的最佳分割方法。     

简缩极化特征值分析的溢油检测

为有效利用简缩极化SAR进行海洋溢油检测,本文基于简缩极化特征值分析,提出了3个用于简缩极化溢油检测的参数,引入了基于简缩极化特征值分解的简缩极化熵Hc(Compact Polarization Entropy)、简缩极化比参数PFc(Compact Polarization Fraction)、简缩极化基准高度PHc(Compact Polarization Pedestal Height)特征进行海洋溢油检测。海表的散射类型主要为低熵散射(小粗糙面发生的Bragg散射),为弱去极化、弱散射过程随机性状态,由于溢油会阻尼海水的Bragg散射,使其熵值变高、呈去极化、强散射过程随机性状态,故简缩极化熵、简缩极化比参数和简缩极化基准高度可以用来检测海洋溢油。本文采用C波段的Radarsat-2、SIR-C/X-SAR数据进行了实验,结果表明:简缩极化熵、简缩极化比参数和简缩极化基准高度能够有效抑制疑似溢油,使海水与疑似溢油差异变小;突出溢油区域,使海水与溢油的可区分性变大。