Analysis of synthetic aperture radar (SAR) images in L-band of short-period internal waves (IWs), and classification of their radar signatures is presented by means of a polarimetric data set from ALOS-PALSAR mission. We choose the polarimetric feature named standard deviation(std) of the co-polarized phase difference (CPD) to identify fundamental differences in SAR signatures of internal waves, and divided them into three different classes, according to their backscattered modulation depths and morphology as well as the std CPD, namely: double-signed, single-negative, and single-positive signatures, for IW normalized image transects that display, respectively, signatures in the form of bright/dark, dark, and bright bands that correspond to positive/negative, negative, or positive variations of radar backscatter. These radar power types of signatures have a counterpart in the std CPD normalized transects, and in this paper we discuss those correlations and decorrelations. We focus in the single-negative type of signature, that is dark bands on gray background, and show that the std CPD is greatly enhanced over the troughs and rear slopes of those IWs. It is suggested that such behavior is consistent with the presence of surface slicks owing to enhanced surfactant concentration. Furthermore, those single-negative SAR signatures appear at locations where and when biological productivity is enhanced. It is found that the modulation depths associated to the std CPD is higher than the one associated to the HH-polarized radar backscatter for single-negative signatures propagating in the range direction, while the reverse occurs for the other types of signatures.

安达曼海是内波频繁发生的海区之一,对其内波的研究是当今海洋研究的热点。本文利用2013—2016年间覆盖整个安达曼海的3 000多幅Terra/Aqua MODIS、GF-1、Landsat-8、Sentinel-1 等卫星遥感图像,从中提取和解译了内波波列线和波向信息,得到安达曼海海洋内波的时间分布特征,并绘制了内波空间分布图。结果表明,安达曼海及其邻近海域内波主要出现在4个区域:苏门答腊岛以北海域、安达曼海中部海域、安达曼海北部海域以及尼科巴群岛以西海域,尺度较大的内波主要分布在苏门答腊岛以北海域和安达曼海中部海域。在时间分布上,2013—2016年间安达曼海内波的年发生次数相近;在热季、雨季及冬季遥感都能观测到内波的发生;2-4月遥感观测到的内波最多,其次为8、9月,7月遥感观测到的内波较少,这可能是由于雨季光学影像受云影响,安达曼海海域晴空影像过少造成,还需要借助更多的遥感影像进一步证明。在波向上,安达曼海多数内波向岸传播,在苏门答腊岛北部、安达曼海中部海域,内波向东或向东南传播;在安达曼群岛东部,内波向东传播,传播一定距离后与海底地形交互作用,一部分继续向前传播,一部分产生反射,向西南方向传播至安达曼群岛;在尼科巴群岛以西海域,内波由尼科巴群岛向孟加拉湾传播。

Internal waves occur frequently in Andaman Sea, and the study of its internal waves is the hotspot of ocean research today. In this study, more than 3 000 Terra/Aqua MODIS, GF-1, Landsat-8 and Sentinel-1 satellite remote sensing images covering the entire Andaman Sea from 2013 to 2016 were analyzed to obtain spatial and temporal distribution and wave direction feature of internal waves in the Andaman Sea. The statistical result shows that internal waves can be observed by the remote sensing mostly in four areas: north of Sumatra Island, middle of the Andaman Sea, north of the Andaman Sea and west of Nicobar Islands. The internal waves with a relative large spatial scale in northern Sumatra and middle of the Andaman Sea are more active in the Andaman Sea. No internal waves in the deep sea and the gradually changing bathymetric gradients areas of the Andaman Sea have been observed by the remote sensing. The distribution of observed internal waves is similar each year from 2013 to 2016 in the Andaman Sea. Internal waves can be observed in the Andaman Sea all through the year, most in February to April (hot season), next in August and September (rainy season), A minimum number of internal waves are observed in June, however, this may be caused by the small amount of clear optical images in rainy season in the Andaman Sea, it needs to further prove by using more remote sensing images. From the directions distribution, most internal waves propagate to the coast. Internal waves propagate eastward or southeastward in north of Sumatra Island and middle of the Andaman Sea, southeastward or southwestward in eastern Andaman Island. In addition, we found some internal waves propagate from the Andaman Sea to the Bay of Bengal.

\n Ocean submesoscale (1–100 km) processes and their substantial impact on Earth’s climate system have been increasingly emphasized in recent decades by high-resolution numerical models and regional observations\n 1–11\n. However, the dynamics and energy associated with these processes, including submesoscale eddies and nonlinear internal waves, have never been observed from a global perspective. Where, when and how much do these submesoscale processes contribute to the large-scale ocean circulation and climate system? Here we show data from the recently launched Surface Water and Ocean Topography (SWOT) satellite\n 12\n that not only confirm the characteristics of submesoscale eddies and waves but also suggest that their potential impacts on ocean energetics, the marine ecosystem, atmospheric weather and Earth’s climate system are much larger than anticipated. SWOT ushers in a new era of global ocean observing, placing submesoscale ocean dynamics as a critical element of the Earth’s climate system.\n

Tropical cyclones (TCs) are associated with severe weather phenomena, making accurate wind field retrieval crucial for TC monitoring. SAR’s high-resolution imaging capability provides detailed information for TC observation, and wind speed calculations require wind direction as prior information. Therefore, utilizing SAR images to retrieve TC wind fields is of significant importance. This study introduces a novel approach for retrieving wind direction from SAR images of TCs through the classification of TC sub-images. The method utilizes a transfer learning-based Inception V3 model to identify wind streaks (WSs) and rain bands in SAR images under TC conditions. For sub-images containing WSs, the Mexican-hat wavelet transform is applied, while for sub-images containing rain bands, an edge detection technique is used to locate the center of the TC eye and subsequently the tangent to the spiral rain bands is employed to determine the wind direction associated with the rain bands. Wind direction retrieval from 10 SAR TC images showed an RMSD of 19.52° and a correlation coefficient of 0.96 when compared with ECMWF and HRD observation wind directions, demonstrating satisfactory consistency and providing highly accurate TC wind directions. These results confirm the method’s potential applications in TC wind direction retrieval.

This article describes the evaluation of a C-band geophysical model function called C-band model 5.N (CMOD5.N). It is used to provide an empirical relation between backscatter as sensed by the spaceborne European Remote Sensing Satellite-2 (ERS-2) and Advanced Scatterometer (ASCAT) scatterometers and equivalent neutral ocean vector wind at 10-m height (neutral surface wind) as function of scatterometer incidence angle.

该文基于美国国家浮标资料中心(National Data Buoy Center,NDBC) 浮标观测数据对哨兵一号搭载的合成孔径雷达(synthetic aperture radar,SAR) 反演风速数据进行精度分析,并利用BP神经网络 (back propagation neural network) 对SAR反演风速的偏差进行校正;同时针对环境要素、BP神经网络训练输入的样本量以及神经网络结构参数设计了敏感性试验;最后将SAR标量风场数据转换为用u、v矢量表示的风场数据,并对u向风和v向风分别进行了精度分析和校正。实验结果表明:SAR反演风速相较于浮标观测数据出现了低估现象;经过BP神经网络校正后,SAR反演风速数据的精度得到了改善,风速的平均偏差绝对值从0.78 m/s下降到0.04 m/s,均方根误差从1.98 m/s下降到了1.77 m/s;敏感性试验表明输入质量较差的环境要素数据时BP神经网络的校正效果有所下降,而增加训练集样本量能改善校正效果;将标量风场数据转换为u、v矢量风场数据后的校正结果也显示BP神经网络具有较好的校正效果。

An accuracy analysis of wind speed data retrieved from Sentinel-1 synthetic aperture radar (SAR) was conducted based on buoy observations from the National Data Buoy Center (NDBC). A back propagation (BP) neural network was utilized to correct the deviation in the SAR-derived wind speeds. Sensitivity experiments were designed for environmental factors, the number of training samples for BP neural network input, and neural network structure parameters. Finally, the SAR wind field data were converted into u and v vector wind data, and the accuracy analysis and correction were performed separately for u and v wind components. The experiment finds that the SAR-derived wind speed is underestimated compared to the buoy data. After calibration using BP neural network, the accuracy of SAR-derived wind speed data is improved, and the absolute value of bias of wind speed decreases from 0.78 m/s to 0.04 m/s, the RMSE of wind speed decreases from 1.98 m/s to 1.77 m/s. The sensitivity experiments suggest that low quality environmental factors input data will decrease the calibration effect of BP neural network, and increasing the sample size of the training set can improve that. The calibration results of converted u and v vector wind field data also show that the BP neural network has good correction effect.

Synthetic Aperture Radar (SAR) imagery presents significant advantages for observing ocean surface winds owing to its high spatial resolution and low sensitivity to extreme weather conditions. Nevertheless, signal noise poses a challenge, hindering precise wind retrieval from SAR imagery. Moreover, traditional geophysical model functions (GMFs) often falter, particularly in accurately estimating high wind speeds, notably during extreme weather phenomena like tropical cyclones (TCs). To address these limitations, this study proposes a novel hybrid model, CMOD-Diffusion, which integrates the strengths of GMFs with data-driven deep learning methods, thereby achieving enhanced accuracy and robustness in wind retrieval. Based on the coarse estimation of wind speed by the traditional GMF CMOD5.N, we introduce the recently developed data-driven method Denoising Diffusion Probabilistic Model (DDPM). It transforms an image from one domain to another domain by gradually adding Gaussian noise, thus achieving denoising and image synthesis. By introducing the DDPM, the noise from the observed normalized radar cross-section (NRCS) and the residual of the GMF methods can be largely compensated. Specifically, for wind speeds within the low-to-medium range, a DDPM is employed before proceeding to another CMOD iteration to recalibrate the observed NRCS. Conversely, a posterior-placed DDPM is applied after CMOD to reconstruct high-wind-speed regions or TC-affected areas, with the prior information from regions characterized by low wind speeds and recalibrated NRCS values. The efficacy of the proposed model is evaluated by using Sentinel-1 SAR imagery in vertical–vertical (VV) polarization, collocated with data from the European Centre for Medium-Range Weather Forecasts (ECMWF). Experimental results based on validation sets demonstrate significant improvements over CMOD5.N, particularly in low-to-medium wind speed regions, with the Structural Similarity Index (SSIM) increasing from 0.76 to 0.98 and the Root Mean Square Error (RMSE) decreasing from 1.98 to 0.63. Across the entire wind field, including regions with high wind speeds, the validation data obtained through the proposed method exhibit an RMSE of 2.39 m/s, with a correlation coefficient of 0.979.

Velocity estimation of ocean surface currents is of great significance in the fields of the fishery, shipping, sewage discharge, and military affairs. Over the last decade, along-track interferometric synthetic aperture radar (along-track InSAR) has been demonstrated to be one of the important instruments for large-area and high-resolution ocean surface current velocity estimation. The calculation method of the traditional ocean surface current velocity, as influenced by the large-scale wave orbital velocity and the Bragg wave phase velocity, cannot easily separate the current velocity, characterized by large error and low efficiency. In this paper, a novel velocity estimation method of ocean surface currents is proposed based on Conditional Generative Adversarial Networks (CGANs). The main processing steps are as follows: firstly, the known ocean surface current field diagrams and their corresponding interferometric phase diagrams are constructed as the training dataset; secondly, the estimation model of the ocean surface current field is constructed based on the pix2pix algorithm and trained by the training dataset; finally, the interferometric phase diagrams in the test dataset are input into the trained model. In the simulation experiment, processing results of the proposed method are compared with those of traditional ocean surface current velocity estimation methods, which demonstrate the efficiency and effectiveness of the novel method.

In this paper, an end-to-end system framework is proposed for the Digital Twin study of spaceborne ATI-SAR ocean current velocity inversion. Within this framework, a fitting inversion approach is proposed to enhance the conventional spaceborne ATI-SAR ocean current velocity inversion algorithm. Consequently, the issue of possible local inversion errors stemming from the mismatch between the traditional spaceborne ATI-SAR inversion algorithm and various dual-antenna configurations is resolved to a certain extent. A simulated spaceborne ATI-SAR system, featuring a dual-antenna configuration comprising a baseline direction perpendicular to the track and a squint angle, is presented to validate the efficacy of the Digital Twin methodology. Under the specified simulation parameters, the average inversion error for the final ocean current velocity is recorded at 0.0084 m/s, showcasing a reduction of 0.0401 m/s compared with the average inversion error prior to optimization.

Accurate measurement of underwater topography in the coastal zone is essential for human marine activities, and the synthetic aperture radar (SAR) presents a completely new solution. However, underwater topography detection using a single SAR image is vulnerable to the interference of sea state and sensor noise, which reduces the detection accuracy. A new underwater topography detection method based on multi-source SAR (MSSTD) was proposed in this study to improve the detection precision. GF-3, Sentinel-1, ALOS PALSAR, and ENVISAT ASAR data were used to verify the sea area of Dazhou Island. The detection result was in good agreement with the chart data (MAE of 2.9 m and correlation coefficient of 0.93), and the detection accuracy was improved over that of a single SAR image. GF-3 image with 3 m spatial resolution performed best in bathymetry among the four SAR images. Additionally, the resolution of the SAR image had greater influence on bathymetry compared with polarization and radar band. The ability of MSSTD has been proved in our work. Collaborative multi-source satellite observation is a feasible and effective scheme in marine research, but its application potential in underwater topography detection still requires further exploration.

The global ocean covers 71% of Earth's surface, yet the seafloor is poorly charted compared with land, the Moon, Mars, and Venus. Traditional ocean mapping uses ship-based soundings and nadir satellite radar altimetry-one limited in spatial coverage and the other in spatial resolution. The joint NASA-CNES (Centre National d'Etudes Spatiales) Surface Water and Ocean Topography (SWOT) mission uses phase-coherent, wide-swath radar altimetry to measure ocean surface heights at high precision. We show that 1 year of SWOT data offers more detailed information than 30 years of satellite nadir altimetry in marine gravity, enabling the detection of intricate seafloor structures at 8-kilometer spatial resolution. With the mission still ongoing, SWOT promises critical insights for bathymetric charting, tectonic plate reconstruction, underwater navigation, and deep ocean mixing.

. A new Sentinel-1 image-based sea ice classification\nalgorithm using a machine-learning-based model trained in a semi-automated\nmanner is proposed to support daily ice charting. Previous studies mostly\nrely on manual work in selecting training and validation data. We show that\nthe readily available ice charts from the operational ice services can\nreduce the amount of manual work in preparation of large amounts of\ntraining/testing data. Furthermore, they can feed highly reliable data to\nthe trainer by indirectly exploiting the best ability of the sea ice experts\nworking at the operational ice services. The proposed scheme has two phases:\ntraining and operational. Both phases start from the removal of thermal,\nscalloping, and textural noise from Sentinel-1 data and calculation of grey\nlevel co-occurrence matrix and Haralick texture features in a sliding\nwindow. In the training phase, the weekly ice charts are reprojected into\nthe SAR image geometry. A random forest classifier is trained with the\ntexture features on input and labels from the rasterized ice charts on\noutput. Then, the trained classifier is directly applied to the texture\nfeatures from Sentinel-1 images operationally. Test results from the two\ndatasets spanning winter (January–March) and summer (June–August) seasons acquired\nover the Fram Strait and the Barents Sea showed that the classifier is\ncapable of retrieving three generalized cover types (open water, mixed\nfirst-year ice, old ice) with overall accuracies of 87 % and 67 % in\nwinter and summer seasons, respectively. For the summer season, the classifier\nfailed in distinguishing mixed first-year ice from old ice with accuracy of\nonly 12 %; however, it performed rather like an ice–water discriminator\nwith high accuracy of 98 % as the misclassification between the mixed\nfirst-year ice and old ice was between them. The accuracy for five cover\ntypes (open water, new ice, young ice, first-year ice, old ice) in the winter\nseason was 60 %. The errors are attributed both to incorrect manual\nclassification on the ice charts and to the semi-automated algorithm.\nFinally, we demonstrate the potential for near-real-time service of the ice\nmap using daily mosaicked Sentinel-1 images.

本文提出了一种利用全极化SAR数据的极化特征获取海冰密集度算法。首先,对全极化SAR数据进行多视化及滤波等预处理,以获取相干矩阵与协方差矩阵;其次,通过相干矩阵与协方差矩阵获取若干极化特征,对这些极化特征进行相关性与冗余性分析,构建最优特征空间;然后,将最优特征空间作为输入量放入神经网络分类器中,得到海冰分类结果;最后,根据海冰分类结果提取海冰密集度。选用拉布拉多南部海域2景全极化Radarsat-2影像获取海冰密集度,与业务化海冰密集度产品ASI-3125进行对比研究。本文算法结果与ASI-3125海冰密集度分布趋势基本一致,总体上略大于ASI-3125海冰密集度,标准差值分布为3.46%和6.82%,说明利用高分辨率全极化SAR数据在监测边缘区域小尺寸破碎海冰方面具有优势。

This paper proposes an algorithm to obtain sea ice concetration(SIC) from fully polarimetric SAR data based on polarization features. Firstly, multilookprocess and filtering are performed on the fully polarimetric SAR data to obtain the coherence matrix and covariance matrix. Secondly, a number of polarization features are obtained through the coherence matrix and covariance matrix, and the correlation and redundancy analysis of these polarization features is carried out to construct the optimal feature space.Then, put the optimal feature space as input into the neural network classifier to obtain the SIC result. Finally, extract the sea ice concentration according to the SIC result. In this paper, two fully polarimetric Radarsat-2 images in the southern waters of Labrador are used to obtain the SIC. Compared with the commercial the SIC product of ASI-3125, the algorithm results of this paper are basically consistent with the distribution trend of the SIC product of ASI-3125,and generally slightly larger than the SIC product of ASI-3125. The standard deviation distributions are 3.46% and 6.82%, indicating that the use of high-resolution fully polarimetric SAR data has advantages in monitoring small-sized broken sea ice in the marginal area.

Utilize 7 Arctic SAR images and level ice thickness from 6th Arctic Survey，calculate texture feature through gray level co-occurrence matrix(GLCM)，confirm suitable GLCM parameters for thickness retrieval，analyze the relationship between sea-ice thickness and texture feature，validate the possibility of sea-ice thickness retrieval from texture feature. Then confirm fitting equation depending on the most suitable texture feature. When validated，the sea-ice thickness retrieval from the empirical equation agrees well with the in-situ data，the average relative error is 13.7%. This value is smaller compared with the commonly used method that only depend on backscattering coefficient，confirm the role of texture feature in sea-ice thickness retrieval.

Oil leaks onto water surfaces from big tankers, ships, and pipeline cracks cause considerable damage and harm to the marine environment. Synthetic Aperture Radar (SAR) images provide an approximate representation for target scenes, including sea and land surfaces, ships, oil spills, and look-alikes. Detection and segmentation of oil spills from SAR images are crucial to aid in leak cleanups and protecting the environment. This paper introduces a two-stage deep-learning framework for the identification of oil spill occurrences based on a highly unbalanced dataset. The first stage classifies patches based on the percentage of oil spill pixels using a novel 23-layer Convolutional Neural Network. In contrast, the second stage performs semantic segmentation using a five-stage U-Net structure. The generalized Dice loss is minimized to account for the reduced oil spill representation in the patches. The results of this study are very promising and provide a comparable improved precision and Dice score compared to related work.

为研究利用紧缩极化SAR代替全极化SAR进行海洋溢油检测的可行性,以及不同极化参数对溢油检测准确率的影响,本文利用卷积神经网络(CNN)的SAR溢油检测算法,对全极化模式及由全极化构造的紧缩极化SAR数据分别提取极化参数,研究其对于溢油分类准确率的影响;同时对比不同SAR数据预处理步骤对溢油检测精度的影响。研究结果表明,对于预处理步骤,线性拉伸方法能够有效提升溢油检测的准确率;在极化参数选择上,极化参数极化熵H在全极化与紧缩极化模式上都取得最高的分类准确率,分别为0.972和0.978。该研究结果证明了利用紧缩极化SAR代替全极化SAR进行溢油检测的可行性,在溢油检测方面具有较好的应用潜力。

To investigate the feasibility of using compact polarimetric synthetic aperture radar (SAR) as an alternative to fully polarimetric SAR for oil spill detection and to determine the impact of different polarization parameters on the accuracy of oil spill detection. To this end,a SAR oil spill detection algorithm based on convolutional neural networks (CNN) is employed. This algorithm extracts polarization parameters from both fully polarimetric and derived compact polarimetric SAR data to study their impact on the classification accuracy of oil spills. Furthermore,the impact of different SAR data preprocessing steps on the accuracy of oil spill detection is evaluated. The results demonstrate that the linear stretching method can effectively enhance the accuracy of oil spill detection. Concerning the selection of polarization parameters,the polarization entropy <i>H</i> achieved the highest classification accuracy in both fully polarimetric and compact polarimetric modes,with a classification accuracy of 0.972 for fully polarimetric and 0.978 for compact polarimetric. This demonstrates the potential of using compact polarimetric SAR for oil spill detection and its promising application prospects.

Compact polarimetric synthetic aperture radar (CP SAR), as a new technique or observation system, has attracted much attention in recent years. Compared with quad-polarization SAR (QP SAR), CP SAR provides an observation with a wider swath, while, compared with linear dual-polarization SAR, retains more polarization information in observations. These characteristics make CP SAR a useful tool in marine environmental applications. Previous studies showed the potential of CP SAR images for ship detection. However, false alarms, caused by ocean clutter and the lack of detailed information about ships, largely hinder traditional methods from feature selection for ship discrimination. In this paper, a segmentation method designed specifically for ship detection from CP SAR images is proposed. The pixel-wise detection is based on a fully convolutional network (i.e., U-Net). In particular, three classes (ship, land, and sea) were considered in the classification scheme. To extract features, a series of down-samplings with several convolutions were employed. Then, to generate classifications, deep semantic and shallow high-resolution features were used in up-sampling. Experiments on several CP SAR images simulated from Gaofen-3 QP SAR images demonstrate the effectiveness of the proposed method. Compared with Faster RCNN (region-based convolutional neural network), which is considered a popular and effective deep learning network for object detection, the newly proposed method, with precision and recall greater than 90% and a F1 score of 0.912, performs better at ship detection. Additionally, findings verify the advantages of the CP configuration compared with single polarization and linear dual-polarization.

The identifying features of ship wakes in synthetic aperture radar (SAR) remote sensing images are of great importance for detecting ships and for extracting ship motion parameters. A statistical analysis was conducted on the identifying features of ship wakes in SAR images in the Yellow Sea. In this study, 1091 ship wake sub-images were selected from 327 SAR images in the Yellow Sea near Qingdao. Analysis of the identifying features of ship wakes in SAR images revealed that both turbulent wakes and Kelvin wakes account for the majority of ship wakes, with turbulent wakes occurring approximately four times as frequently as Kelvin wakes. Narrow-V wakes and internal wave wakes were comparatively rare, which is due to the peculiarities of the radar system parameters and marine environments required to observe these wakes. Additionally, we extracted ship motion parameters from four types of ship wakes in the SAR images. Specifically, internal wave wakes in SAR images in the Yellow Sea were also used to extract ship motion parameters. Validation of the extracted parameters indicated that the extraction of these parameters from ship wakes is a viable and accurate approach for the acquisition of ship motion parameters. These results provide a solid foundation for the commercialization of SAR-based technologies for detecting ships and extracting ship motion parameters.

. Offshore wind energy is at the advent of a massive global expansion. To investigate the development of the offshore wind energy sector, optimal offshore wind farm locations, or the impact of offshore wind farm projects, a freely accessible spatiotemporal data set of offshore wind energy infrastructure is necessary. With free and direct access to such data, it is more likely that all stakeholders who operate in marine and coastal environments will become involved in the upcoming massive expansion of offshore wind farms. To that end, we introduce the DeepOWT (Deep-learning-derived Offshore Wind Turbines) data set (available at https://doi.org/10.5281/zenodo.5933967, Hoeser and Kuenzer, 2022b), which provides 9941 offshore wind energy infrastructure locations along with their deployment stages on a global scale. DeepOWT is based on freely accessible Earth observation data from the Sentinel-1 radar mission. The offshore wind energy infrastructure locations were derived by applying deep-learning-based object detection with two cascading convolutional neural networks (CNNs) to search the entire Sentinel-1 archive on a global scale. The two successive CNNs have previously been optimised solely on synthetic training examples to detect the offshore wind energy infrastructures in real-world imagery. With subsequent temporal analysis of the radar signal at the detected locations, the DeepOWT data set reports the deployment stages of each infrastructure with a quarterly frequency from July 2016 until June 2021. The spatiotemporal information is compiled in a ready-to-use geographic information system (GIS) format to make the usability of the data set as accessible as possible.

Offshore wind farms are widely adopted by coastal countries to obtain clean and green energy; their environmental impact has gained an increasing amount of attention. Although offshore wind farm datasets are commercially available via energy industries, records of the exact spatial distribution of individual wind turbines and their construction trajectories are rather incomplete, especially at the global level. Here, we construct a global remote sensing-based offshore wind turbine (OWT) database derived from Sentinel-1 synthetic aperture radar (SAR) time-series images from 2015 to 2019. We developed a percentile-based yearly SAR image collection reduction and autoadaptive threshold algorithm in the Google Earth Engine platform to identify the spatiotemporal distribution of global OWTs. By 2019, 6,924 wind turbines were constructed in 14 coastal nations. An algorithm performance analysis and validation were performed, and the extraction accuracies exceeded 99% using an independent validation dataset. This dataset could further our understanding of the environmental impact of OWTs and support effective marine spatial planning for sustainable development.© 2021. The Author(s).

针对海上油气平台信息不足的问题,开展多源卫星遥感的油气平台识别方法研究。基于Landsat-8光学遥感影像(2018—2021年)应用阈值分割法、K-means分类法和最大似然分类法分别识别出渤海海域油气平台136座、166座和113座;基于Sentinel-1 SAR影像(2018—2021年)应用阈值分割法识别出油气平台338座;对上述结果进行决策级融合,识别出渤海油气平台428座。利用ZY-3高分辨率影像对融合方法的识别结果进行验证,结果显示识别油气平台的正确率达到85.2%,错判率、漏判率分别为10.9%和3.9%;油气平台位置与相关文献和公开资料一致。研究结果表明,决策级融合方法能够实现海上油气平台的有效判别,具有推广、应用价值。

To solve the problem of insufficient information of offshore oil and gas platform, the method of oil and gas platform identification based on multi-source satellite remote sensing was studied. Based on Landsat-8 remote sensing images of the Bohai Sea (2018-2021), 136, 166 and 113 oil and gas platforms in the Bohai Sea were identified by threshold segmentation, K-means unsupervised algorithm and maximum likelihood classification, respectively. Based on Sentinel-1 SAR images (2018-2021), 338 oil and gas platforms were identified by threshold segmentation method. Based on the decision level fusion of the above results, 428 oil and gas platforms in the Bohai Sea were identified. The ZY-3 high-resolution images were used to verify the identification results of the fusion method. The results showed that the accuracy of the identified oil and gas platforms reached 85.2%, and the error rate and miss rate were 10.9% and 3.9%, respectively. The identified oil and gas platform locations are consistent with literature and public data. The research shows that the decision level fusion method can realize the effective identification and extraction of offshore oil and gas platforms, and has the value of popularization and application.

Raft-culture is a way of utilizing water for farming aquatic product. Automatic raft-culture monitoring by remote sensing technique is an important way to control the crop’s growth and implement effective management. This paper presents an automatic pixel-wise raft labeling method based on fully convolutional network (FCN). As rafts are always tiny and neatly arranged in images, traditional FCN method fails to extract the clear boundary and other detailed information. Therefore, a homogeneous convolutional neural network (HCN) is designed, which only consists of convolutions and activations to retain all details. We further design a dual-scale structure (DS-HCN) to integrate higher-level contextual information for accomplishing sea–land segmentation and raft labeling at the same time in a uniform framework. A dataset with Gaofen-1 satellite images was collected to verify the effectiveness of our method. DS-HCN shows a satisfactory performance with a better interpretability and a more accurate labeling result.