Currently, global directional wave spectra are routinely providing by synthetic aperture radar (SAR) on Sentinel-1 in Wave Mode and real aperture radar, i.e., SWIM onboard CFOSAT. However, SAR spectra suffer from azimuth cut-off effects, while SWIM spectra exhibit parasitic peaks and 180° ambiguity. Leveraging the complementary of these two satellite radars for directional spectra merging remains an open scientific challenge. In this letter, we propose a deeply-learned based 2D wave spectrum fusion model aiming to integrate CFOSAT SWIM and Sentinel-1A wave mode Level-2 products. After spatiotemporal collocating, and rejecting suspicious spectra, two-year period (June 2022 to June 2024) of 2D wave spectra triplets (S-1, SWIM, and ERA5) are used for the fusion model training. The deep learning merged spectra show good consistency with ERA5 benchmark regarding 2D spectra, the derived omni - directional energy distribution and the integrated wave parameters. Comparative analysis with ERA5 reanalysis demonstrates that our fused directional spectrum could effectively compensate for high-frequency information loss and spectral distortion in SAR products, meanwhile suppress SWIM's parasitic peaks and resolve directional ambiguity. Qualitatively, compared to ERA5, the Brüning’s correlation coefficient and square error of our fusion results reaches 0.95 and 0.17, significantly outperforming official SAR (0.79 and 0.58) and SWIM (0.83 and 0.52) products. The fusion mitigates SAR’s high-frequency losses and SWIM’s artifacts, enhancing ocean wave spectrum accuracy for improved geophysical applications.

This paper presents a comprehensive mathematical approach to address the multi-player and multi-objective pursuit-evasion games problem, incorporating coalition structure constraints from a cooperation-competition perspective. Social interaction networks are developed to approximate priority communication alliances based on individual preferences, establishing a multi-connected topology and decision space for the games. An N-player variable-sum differential game model, featuring autonomous obstacle avoidance, is formulated by integrating kinematic constraints and the social forces method. Rigorous proofs are provided for the uniqueness of payoff distribution, the stability of alliance structures, and the convergence of many-to-many differential games to Nash Equilibrium. Simulation and experimental results are presented to validate the effectiveness and performance of the proposed method.

Seeking the Pareto-Nash equilibrium in multi-agent, multi-objective general-sum games (MMGSG) poses a significant challenge, particularly in accurately capturing individual preferences and adhering to the fairness principle of the solution. To address this issue, this paper introduces, for the first time, a multi-level distributed fuzzy optimum policy iteration (MDFOPI) method for identifying the Pareto-Nash equilibrium point in MMGSG. This approach is grounded in fuzzy optimal membership degrees, and employs fuzzy measures and$\lambda$-mean classification to construct the coupled multi-objective optimum matrix, utilizing the strategy space as the foundation. The Pareto-Nash equilibrium point is sought through the MDFOPI method, with the multi-objective optimal membership degree matrix used to organize the sampled data and integrate the results of multi-objective evaluations. This work rigorously proves the existence of Nash equilibria in MMGSG and establishes the convergence of the MDFOPI method to a fixed point, specifically a Pareto-Nash equilibrium point. The accuracy and practical applicability of the research findings are verified through simulation experiments.

Marine oil spills pose significant threats to ecosystems and human health, emphasizing the importance of synthetic aperture radar (SAR) images for reliable and all-weather monitoring. However, current methods face two major challenges. The first is data limitations, including insufficient data quantity and noise, such as speckle noise and distortions introduced during preprocessing. The second is look-alike confusions, which pose challenges in distinguishing oil spills from visually similar phenomena. This article introduces TransOilSeg, a novel method designed to address these challenges and enhance oil spill detection performance. TransOilSeg employs a transfer learning component (TLC) to integrate data from diverse geographical regions and varying quality, learning general features from multisource datasets. By leveraging a gradient aggregation algorithm, the model combines features from limited and noisy SAR oil spill (SOS) datasets, transferring data deficiencies. In addition, the adaptive attention hybrid encoder (AAHE) analyzes contextual features and adapts to varying datasets, enabling the model to effectively distinguish oil spills from look-alike phenomena. Comprehensive evaluations across multiple datasets demonstrate the robust generalization capability of TransOilSeg. On the M4D dataset, which includes 1002 training samples, the model achieved a mean intersection over union (mIoU) of 61.38% for oil spill detection and 62.41% for look-alike detection. Furthermore, TransOilSeg maintained strong performance when transferred between datasets with varying levels of noise and distortions, demonstrating its adaptability to challenging conditions. These results highlight its potential as a reliable tool for marine oil spill detection and monitoring.

The integration of cooperation and competition strategies in game theory emphasizes the systematic nature of strategy spaces and group interactions, forming the basis for achieving win-win scenarios. This is particularly crucial under coalition constraints and incomplete information. Addressing these challenges, this article introduces a comprehensive mathematical method for policy formation using co-opetition topological networks. This method enables autonomous decision-making and game equilibrium in group decision scenarios, considering individual preferences amidst constraints like incomplete information and alliance limitations. Leveraging the complementary entropy theorem on superiority, inferiority, and fuzzy measures, we propose a cognitive model for information interaction and attribute fusion. Utilizing the ordered weighted averaging operator and average tree solutions aids in identifying optimal alliance structures. We subsequently discuss evaluating missing information to complete the topological network. Updating the cognitive model and value function, we develop a Gaussian oscillation heuristic algorithm to explore alliance and component strategy spaces. Simulation results are provided and analyzed to illustrate the performance and effectiveness of our approach.

This paper evaluates the Fengyun-3E (FY-3E) Global Navigation Satellite System Occultation Sounder II (GNOS-II) global wind speed product using measurements from 91 National Data Buoy Center (NDBC) buoys in offshore regions and 48 Tropical Atmosphere Ocean/Triangle Trans-Ocean Buoy Network (TAO/TRITON) buoys in the open ocean. The study underscores the differences in wind speed accuracy between offshore and open ocean regions, supporting the application of the GNOS-II system in various maritime environments. The findings reveal that nearshore wind speed measurements have a lower accuracy compared to those in the open ocean. Specifically, the Root Mean Square Error (RMSE) for the BeiDou Navigation Satellite System (BDS) product against NDBC buoy data is 2.494 m/s, while for the Global Positioning System (GPS) product it is 2.128 m/s. Against TAO/TRITON buoy data, the RMSE for the BDS product is 1.831 m/s, and for the GPS product, it is 1.918 m/s. These results indicate a need for further enhancement of offshore wind speed measurements to meet application requirements.

Ship target search tasks aim to match specific ships across two or more satellite images. Like pedestrian and vehicle re-identification tasks in computer vision, accurate ship re-identification encounters challenges, including subtle differences between ships of the same type and substantial intra-instance variations due to satellite angle of view and spectral differences. To tackle these challenges, this paper introduces a deep learning-based two-branch framework for ship target search, integrating ship detection and re-identification tasks. One branch extracts the target ship features while the other captures the search region features. These features are then fused through a dedicated layer, and the final output is derived from the keypoint detection header. A new dataset was curated using Sentinel-2 and Gaofen-1 satellite data. Experimental results validate the robustness of the proposed method, achieving an accuracy of 94.37% on the new dataset. Our method’s scalability has been validated through experiments using CBERS-04 and Gaofen-6 satellite data.

Synthetic-aperture radar (SAR) plays a crucial role in monitoring the fine structure of tropical cyclones, but its effectiveness is constrained by limitations such as signal degradation and saturation. To address this challenge, we proposed a transfer learning-based generative adversarial network (GAN) framework with a dilated convolution and attention mechanism for reconstructing inner-core high winds from SAR images. We have employed the principles of transfer learning to adapt pre-trained models developed by the HWRF (Hurricane Weather Research and Forecasting model) winds to SAR images during tropical cyclone events for reconstruction. The proposed model can effectively capture the relationship between features in the low-precision areas and global features from SAR images, facilitating tropical cyclone wind speed reconstruction. The utilization of Global Precipitation Measurement (GPM) Level 3 rainfall data facilitates the identification of rainfall regions in 89 SAR images obtained from Radarsat-2 and Sentinel-1A/B missions. Comparison with Stepped Frequency Microwave Radiometer (SFMR) data reveals that the model exhibits a bias of –0.69 m/s, an RMSE of 4.08 m/s, and anRvalue of 0.91 under heavy rainfall conditions (>7.62 mm/hr). Remarkably, the GAN model exhibits excellent performance compared with measurements from the Soil Moisture Active Passive (SMAP) L-band radiometer, achieving an RMSE of 3.78 m/s. Our findings indicate that deep learning technology holds significant promise for the reconstruction and monitoring of tropical cyclones through the utilization of SAR imagery.

POI tagging aims to annotate a point of interest (POI) with some informative tags, which facilitates many services related to POIs, including search, recommendation, and so on. Most of the existing solutions neglect the significance of POI images and seldom fuse the textual and visual features of POIs, resulting in suboptimal tagging performance. In this paper, we propose a novel M ulti-M odal M odel for P OI T agging, namely M3PT, which achieves enhanced POI tagging through fusing the target POI's textual and visual features, and the precise matching between the multi-modal representations. Specifically, we first devise a domain-adaptive image encoder (DIE) to obtain the image embeddings aligned to their gold tags' semantics. Then, in M3PT's text-image fusion module (TIF), the textual and visual representations are fully fused into the POIs' content embeddings for the subsequent matching. In addition, we adopt a contrastive learning strategy to further bridge the gap between the representations of different modalities. To evaluate the tagging models' performance, we have constructed two high-quality POI tagging datasets from the real-world business scenario of Ali Fliggy. Upon the datasets, we conducted the extensive experiments to demonstrate our model's advantage over the baselines of uni-modality and multi-modality, and verify the effectiveness of important components in M3PT, including DIE, TIF and the contrastive learning strategy.

In this study, we propose a method for wind speed retrieval using a random forest (RF) algorithm for Cyclone Global Navigation Satellite System (CYGNSS) data. We first compared CYGNSS data with Soil Moisture Active Passive (SMAP) data and found a certain deviation in the CYGNSS ”young sea, limited fetch” (YSLF) data product for high winds. Then, we used SMAP as the ”ground truth” to train an RF model and applied it to the wind speed retrieval of CYGNSS data. The experimental results show that using the RF algorithm for wind speed retrieval can eliminate noise in the CYGNSS YSLF wind speed data and improve retrieval accuracy. In addition, we explored the impact of different input parameter combinations on model performance and found that using an 11-parameter model in CYGNSS wind speed retrieval can achieve optimal performance. This can provide valuable reference for rapid near-real-time retrieval of tropical cyclones using CYGNSS.

Although cross-polarized synthetic aperture radar (SAR) images play a crucial role in wind speed retrieval under extreme weather conditions, the retrieved wind speeds are susceptible to thermal noise. However, there is still a lack of research on how thermal noise affects the accuracy of wind speed retrieval. To address this issue, this paper proposes a strategy to quantitatively examine the impact of thermal noise on wind speed retrieval, using 910 Sentinel-1 cross-polarized SAR images acquired in Extra-Wide Swath (EW) mode. The thermal noise and wind speeds of these images range from -35 dB to -22.5 dB and from 5 m/s to 70 m/s, respectively. By considering the wind speed retrieved from dual-polarized SAR as a reference, the study reveals that higher levels of thermal noise result in increased uncertainty in the retrieved wind speeds using cross-polarized SAR images. Additionally, the error of wind speed retrieval from cross-polarized SAR rapidly decreases as wind speed increases, eventually converging to a stable level. Notably, as thermal noise levels decrease to less than -30 dB for wind speeds between 5 m/s and 25 m/s, the root-mean-square error (RMSE) associated with wind speed retrieval through cross-polarized SAR imagery experiences a rapid decline, with a 94% reduction in RMSE, which then stabilizes. This implies that when the thermal noise level falls below -30 dB, wind speed retrieval can be directly conducted using cross-polarized imagery for wind speeds exceeding those typical of a tropical storm, yielding comparable results to dual-polarized imagery. The findings of this study provide valuable insights for advancing wind speed retrieval algorithms and hold significant reference value for the design of the next-generation radar instruments that will incorporate the cross-polarized channel.

Characterizing the uncertainties in buoy ocean wave records is critical not only for understanding the limitations ofin situwave measurements, but also for interpreting the implied accuracies of the remotely sensed products in which these buoy data are used as validation references. This letter preliminarily assesses the error of long-period swell heights (Hss) representing specific directional wave partition energy observed from deep-water buoys moored in the northeast Pacific. We propose a buoyHsserror estimation method by combining dual and triple collocation using data derived from buoys, two kinds of space-borne radars and numerical simulations. Compared to traditional methods, the proposed approach can reveal “absolute” errors (with respect to the underlying truth) from buoyHss, accepting and then confirming that swell heights from buoy, satellite and model are all uncertain. This study simultaneously employs ocean swell products derived from synthetic/real aperture radars (Sentinel-1A/B and CFOSAT/SWIM) and WaveWatch III® ocean wave model hindcasts to diagnose the accuracy of theHssvalues observed by buoys of National Data Buoy Center (NDBC) and Coastal Data Information Program (CDIP) during the period from July 2019 to October 2021. We quantify that the NDBC’s 3-m heave-pitch-roll buoy (CDIP’s Waverider buoy) recordedHsshave root-mean-square error of 0.17 m (0.12 m), or have about 10.65% (7.06%) uncertainty relative to the meanHssvalue (approximately 1.6 m). Our findings imply that the reference value uncertainties should be taken into account when understanding direct satelliteHssvalidation against buoyin situ.

Effective and efficient classification of synthetic aperture radar (SAR) images represents an important step toward image interpretation and knowledge discovery. Keys to classification performance include feature extraction and availability of class labels for training. Generative adversarial networks (GANs) are a recently advanced powerful framework to offer both automatic feature extraction and semisupervised and unsupervised learning. For its known drawbacks such as mode collapse and training instability, noteworthy improvements include the deep convolutional GANs (DCGANs). In the letter, we proposed a new approach, which further advances DCGANs with autoencoder (AE) for intermediate feature extraction and reconstruction, in combination with multiclassifier (MC) for better context treatment. The resulting improvement on training stability and mode preservation over DCGAN is clearly demonstrated by the classification results across multifrequency bands (L-, C-, and X-bands).

The surface wave investigation and monitoring (SWIM) aboard the China-France Oceanography Satellite (CFOSAT), a pioneer conically scanning wave spectrometer, was successfully launched on October 29, 2018. Its innovative configuration composed of one nadir and five rotating near-nadir beams is designed to simultaneously observe the directional wave spectrum at a global scale. In this study, we systematically implement the spectral analysis of the radar backscattering with the periodogram technique to obtain the fluctuation spectrum for each azimuth direction. The 2-D fluctuation spectrum of the three spectral beams ($\theta = 6^\circ $, 8°, and 10°) combines all the azimuth directions within one entire rotation of 360°. The case study demonstrates that the wave features (peak wavelength and direction) are roughly consistent between the estimated fluctuation spectrum and the collocated WaveWatch III wave slope spectrum. A marked up-to-downwave asymmetry of the fluctuation spectrum with larger spectral level in the upwave direction for all the three spectral beams is observed. A ratio is defined between the fluctuation spectrum within the [0°, 180°] sector relative to the [180°, 360°] sector. Statistics display that this ratio is greater than 1 when it denotes the up-to-downwave ratio and smaller than 1 for the down-to-upwave ratio. This observed spectrum asymmetry is linked to the asymmetric modulation from upwind to downwind. In addition, we employ such finding to help remove the 180° wave direction ambiguity from a practical point of view. Preliminary results of the direction ambiguity removal display a bias of 41.3°, 40.6°, and 36.7° for the beams. The 10° beam shows slightly better performance compared to the other two beams in terms of bias and standard deviation. This shall lay a strong basis for the operational implementation of such algorithm to resolve the direction ambiguity.

Nowadays, Sentinel-1 (S-1) synthetic aperture radars (SARs) operating in wave mode and the real aperture radar (RAR) called Surface Waves Investigation and Monitoring (SWIM) onboard the China-France Oceanography SATellite (CFOSAT) are the only two kinds of spaceborne radars providing directional ocean wave information globally. To quantify the absolute uncertainties in the swell wave heights of a specific wave system (Hss) observed from these two spaceborne sensors, a triple colocation error model is exploited via WaveWatch III (WW3) wave model hindcasts for the first time. After implementing spatiotemporal collocation, cross-assigning swell partitions, and rejecting suspicious data, a database of the optimal-matched Hss triplets (S-1, SWIM, and WW3) is determined over a one-year period (June 2020 to June 2021). Qualitatively, traditional dual intercomparisons indicate the inconsistency between the Hss from both SAR and RAR radars in terms of systematic biases. Furthermore, the triple collocated error analysis quantitively reveals that, at a global scale, SWIM onboard CFOSAT has the least uncertainty in Hss (~0.2 m root-mean-square error (RMSE) and ~11% scatter index (SI)) compared with S-1 SAR (0.35-0.50 m RMSE and 17% - 26% SI depending on incidence modes) and WW3, under the assumption that the random errors of the three data sources are independent, indicating that the newly launched SWIM instrument is an invaluable resource of directional wave information for the scientific community. The results are discussed with respect to regional error characteristics along with a feasible explanation of error sources. The findings could be helpful for better understanding and synergistically exploiting the Hss datasets from these two spaceborne radars.

Wind-induced oriented textures (WIOTs) are commonly used to retrieve sea surface wind directions from synthetic aperture radar (SAR) images. In this study, we found that WIOTs are also related to sea surface wind speeds (SSWSs). The entropy values in the gray-level cooccurrence matrices (GLCMs) for SAR images containing WIOTs will become steady with increasing distance between pairs of pixels. Furthermore, these steady values of entropy (SVEs) show a clear linear relationship with SSWSs. As a result, an SSWS retrieval model was developed based on this relationship. We used 2222/2223 Sentinel-1 SAR images (wind speed ranges from 5 to 20 m/s) to fit/validate the algorithm. The retrieved SSWSs were compared with the European Centre for Medium-Range Weather Forecast (ECMWF) SSWSs, Cross-Calibrated Multi-Platform (CCMP) SSWSs, and Tropical Atmosphere/Ocean (TAO) buoy measurements, and the root-mean-square differences (RMSDs) were 1.78, 1.70, and 1.78 m/s, respectively. The new model was also tested for SAR images acquired under hurricane conditions. The wind comparisons against stepped-frequency microwave radiometer (SFMR) measurements show an RMSD of 1.28 m/s. Our model’s performance was also tested with the images at different spatial scales in the validation data set. Since the model is based on inherent image patterns, it still works well for SAR images without precise calibration.

The potential mymargin for improving storm surge simulation is demonstrated by using winds derived from ground-based Global Navigation Satellite System Reflectometry (GNSS-R) that uses BeiDou geostationary Earth orbit (GEO) satellite signals. We reconstruct wind fields by blending GNSS-R coastal winds with the European Center for Median Weather Forecasts (ECMWF) reanalysis product. The reconstructed winds agree well with the weather station data collected at Yangjiang in Guangdong, China. The ECMWF winds and the reconstructed winds are used to force a storm surge model off the Chinese coast during typhoon Utor 2013, respectively. The model storm surges forced by the reconstructed winds agree substantially better with tide-gauge observations than those forced by the ECMWF winds. The average error has been reduced by 30.5% from 24.3 cm with the ECMWF winds to 16.9 cm with the reconstructed winds. This letter suggests that GNSS-R coastal winds can have a positive impact on the accuracy of storm surge hindcasting directly and forecasting indirectly by improving the initial conditions.

In this letter, we modify the extended advanced integral equation model (EAIEM) for electromagnetic backscattering and bistatic scattering from rough surfaces with small to moderate heights. We extend the first-order approximation of the error function as introduced in the EAIEM model to the second order, in a hope to be more suitable for large roughness and high frequency. In addition, a new transition model for the reflection coefficient is proposed to make the dependences explicit on the mean surface curvature, frequency, and dielectric constant, whereas making no use of the complementary term, the effect of inadequate evaluation of this term is mitigated. Comparison with POLARSCAT data for backscattering and with European Microwave Signature Laboratory (EMSL) measurements for bistatic scattering demonstrates the validity of the updated model.

With the rapid development of sensor technology, lots of remote sensing data have been collected. It effectively obtains good semantic segmentation performance by extracting feature maps based on multi-modal remote sensing images since extra modal data provides more information. How to make full use of multi-model remote sensing data for semantic segmentation is challenging. Toward this end, we propose a new network called Multi-Stage Fusion and Multi-Source Attention Network ((MS) 2 -Net) for multi-modal remote sensing data segmentation. The multi-stage fusion module fuses complementary information after calibrating the deviation information by filtering the noise from the multi-modal data. Besides, similar feature points are aggregated by the proposed multi-source attention for enhancing the discriminability of features with different modalities. The proposed model is evaluated on publicly available multi-modal remote sensing data sets, and results demonstrate the effectiveness of the proposed method.

Gaofen-4 (GF-4) is the first high-resolution geostationary satellite of China, launched on December 29, 2015. Its visible-near infrared optical sensor is capable of imaging the earth at 50 m resolution, covering a 512 km × 512 km area with a minimum imaging interval of 20 s. More than 300 GF-4 images, taken in four sequences with imaging rates of 36 and 69 s per scene, captured the development of Typhoon Megi-2016-from its peak in the afternoon of September 26, 2016 to its dispersion two days later on September 28. These consecutive images recorded the motion field of the typhoon's top clouds nearly continuously. By using advanced image matching technology, the motion has been estimated for every pixel, at subpixel accuracy from image pairs with 179 and 206 s intervals. The process has generated time series of atmospheric motion vector (AMV) fields at 50 m spatial resolution for the whole imaged area. It is the first time, to our knowledge, that such high-resolution and nearly continuous AMV data of a typhoon system have been produced. The data provide accurate measurements of the typhoon top cloud motion speed and direction, and reveal quantitative details of the motion field spatiotemporal evolution at high altitude. The finding that the high altitude top cloud motion speed is significantly lower than that recorded at low altitude below the planetary boundary layer is of scientific value and further exploitation of the motion data can lead to a better understanding of typhoon dynamics and thus improve cyclone modeling.

This paper assesses different retrieval schemes used for the Chinese Gaofen-3 Synthetic Aperture Radar (GF-3 SAR) co-polarized data. The data consist of 4186 GF-3 data points and collocated wind information from sources including the ASCAT scatterometer, HY2A-SCAT scatterometer, and National Data Buoy Center (NDBC) buoy wind data set. The VV-polarized geophysical model function (GMF) is a CMOD7 model while the HH-polarized GMF is a hybrid of the CMOD7 and PR model. Assessments involve comparisons between SAR-derived and collocated winds in terms of the root-mean-square difference (RMSD) and bias. First, a comparison between the two retrieval schemes for the VV-polarized data clearly shows that the optimal scheme performs better than the classical scheme for wind speed retrieval. Comparisons for HH-polarized data show similar results. These experiments indicate that the wind speed RMSDs for the GF-3 co-polarized data are within 2 m/s when using the optimal scheme. Moreover, the wind direction RMSDs from the two schemes have no significant difference, with values near 20°. Overall, these assessments indicate that the GF-3 co-polarized data are sufficient for operational wind speed retrieval using the optimal scheme. However, wind direction retrieval requires further improvement.

This article presents a novel technique for automatically locating tropical cyclone (TC) centers based on top cloud motions in consecutive geostationary satellite images. The high imaging rate and spatial resolution images of the Gaofen-4 geostationary satellite enable us to derive pixel-wise top cloud motion data of TCs, and from the data, TC spiral centers can be accurately determined based on an entirely different principle from those based on static image features. First, a physical motion field decomposition is proposed to eliminate scene shift and TC migration in the motion data without requiring any auxiliary geolocation data. This decomposition does not generate the artifacts that appear in the results of the previously published motion field decomposition. Then, an algorithm of a motion direction-based index embedded in a pyramid searching structure is fully designed to automatically and effectively locate the TC centers. The test shows that the TC concentric motions are more clearly revealed after the proposed motion field decomposition and the located centers are in good agreement with the cloud pattern centers in a visual sense and also with the best track data sets of four meteorological agencies.

The brightness temperature data acquired during 2012-2015 from the scanning microwave radiometer (SMR), onboard the first Chinese ocean dynamic environment satellite- Haiyang-2A, were matched up with the WindSat Polarimetric Radiometer (WindSat) 0.25° × 0.25° gridded daily sea surface temperature (SST) data. Then, the artificial neural network (ANN) ensemble (ANNE) method implementing the Crogging technique was used to build the SMR SST retrieval algorithm. Different from a regular ANN, an ANNE combines the outputs of its ANN members to generate an algorithm. The developed ANNE algorithm for SMR SST was validated based on the SMR/WindSat data pairs that were not used in the tuning of the algorithm. The SST comparison shows the root mean square (rms) of 1.16 °C for the ANNE algorithm. We further validate the SMR SST products using the in situ measurements from the National Oceanic and Atmospheric Administration iQuam System. The rms of the ANNE algorithm in comparison with the global iQuam SSTs is 1.46 °C. All validations showed that ANNEs were more accurate than the other statistically based SST retrieval algorithms for SMR, and generally had much smaller uncertainties than regular ANNs.

Considering its advantages in dealing with high-dimensional visual input and learning control policies in discrete domain, Deep Q Network (DQN) could be an alternative method of traditional auto-focus means in the future. In this paper, based on Deep Reinforcement Learning, we propose an end-to-end approach that can learn auto-focus policies from visual input and finish at a clear spot automatically. We demonstrate that our method - discretizing the action space with coarse to fine steps and applying DQN is not only a solution to auto-focus but also a general approach towards vision-based control problems. Separate phases of training in virtual and real environments are applied to obtain an effective model. Virtual experiments, which are carried out after the virtual training phase, indicates that our method could achieve 100% accuracy on a certain view with different focus range. Further training on real robots could eliminate the deviation between the simulator and real scenario, leading to reliable performances in real applications.

A gray-level co-occurrence matrix (GLCM)-based method was developed for better texture orientation estimation in remote sensing imagery. A GLCM is essentially the joint probability distribution of gray levels at the position pairs satisfying a specific relative position within an image. We first found that when the relative position is aligned with texture orientation, larger elements of the corresponding GLCM are concentrated diagonally. Then, we developed a new texture orientation estimation method. The method uses the GLCMs of relative positions equally spaced in orientation and distance, and three schemes of these GLCMs are calculated. A GLCM-derived parameter is then defined to quantitatively measure the degree of diagonal concentration of the GLCM elements, and its integral over the variable of relative distance is selected as an indicator to find the dominant texture orientation(s). For testing, we applied the method to 44 selected images containing one or multiple aligned textures. The results show that the method is in good agreement with visual inspections from 45 randomly selected people, and is insensitive to large typical noises and illumination change. In addition, using (any) one GLCM calculation scheme over the others does not significantly affect the results. Finally, the method was applied to sea surface wind direction (SSWD) retrieval from 89 synthetic aperture radar images. In the application test, the developed method achieves better SSWD retrieval accuracy than do the commonly used Fourier transform- and gradient-based methods by 8.13° and 16.09° against the European Centre for Medium-Range Weather Forecast ERA-Interim reanalysis data and 10.21° and 17.31° against the cross-calibrated multiplatform data.

In this letter, we present a business process flow of marine target detection using GF-3 SAR data. It includes 12 kinds of imaging modes and 3 types of polarization. The detection method for the single polarization data is a Double-Parameter Constant False Alarm Ratio (DP-CFAR) algorithm. For double polarization data, a pixel level fusion strategy between two single polarization data is used. For the four polarization data, the method of target detection includes three steps: decomposition, false color composite, and supervised classification. All the data we used are test data of the GF-3 satellite. The results of target detection can be visually checked, and can also be validated by an automatic identification system (AIS), which will be done next month. This is submitted for a special session of “New Developments of Chinese Oceanographic and Meteorological Satellites”.

Surface Wave Investigation and Monitoring (SWIM) will be launched on board the Chinese French Ocean SATellite (CFOSAT) in 2018. This paper proposes a joint method to simultaneously retrieve wave spectra at different scales from spaceborne Synthetic Aperture Radar (SAR) and CFOSAT SWIM wave spectrometer data. The method combines the output from the two different sensors to overcome retrieval limitations that occur in some sea states. The wave spectrometer sensitivity coefficient is estimated using an effective significant wave height (SWH), which is an average of SAR-derived and wave spectrometer-derived SWH. This averaging extends the area of the sea surface sampled by the nadir beam of the wave spectrometer to improve the accuracy of the estimated sensitivity coefficient in inhomogeneous sea states. Wave spectra are then retrieved from SAR data using wave spectrometer-derived spectra as first guess spectra to complement the short waves lost in SAR data retrieval. In addition, the problem of 180° ambiguity in retrieved spectra is overcome using SAR imaginary cross spectra.

With the current and planned scientific missions related to ocean observation, there is a need to understand the relevant scattering physics and sensitivity so as to provide a physical basis for retrieving geophysical information from ocean-scattered signals. Sea spectrum is one of the factors to be determined in the retrieval. Since difference in sea spectrum manifests itself mostly at sea waves of intermediate and large scales, it is desirable to include sea waves of all scales (large, intermediate, and small scales) at the same numerical simulation. In this study, we extend the stochastic second degree iterative algorithm with sparse matrix and Chebyshev approximation (SSD-SM-Cheby) method that we previously developed to the analysis of sensitivity of bistatic scattering pattern upon sea spectrum at both L and C bands at low wind speed. Four popular sea spectra are considered. Results are provided and discussed.

China launched two new spaceborne microwave sensors in August and September 2016. One is the C band multi-polarization high resolution synthetic aperture radar (SAR) on board satellite GF-3. The other is the Ku band wide swath Interferometric Imaging Radar Altimeter (InIRA) on board space laboratory TG-2. This paper gives some preliminary results for the quantitative remote sensing of ocean winds and waves from the GF-3 SAR and the TG-2 InIRA. Comparisons to the ECMWF ERA-Interim reanalysis data show good agreements but more valuable details.

The Interferometric Imaging Radar Altimeter (InIRA) is a new generation radar altimeter developed by China, which combines the function of interferometric radar altimeter and Synthetic Aperture Radar (SAR). In this paper, we explored the retrieval method of wind speed from simulated InIRA echoes. The combined significant wave height (Hs) was first statistically derived from sea surface height (SSH), which was retrieved using the similar method by conventional altimeter. The swell Hs was estimated using swell spectrum extracted from InIRA data. Then the wind wave Hs was estimated using combined and swell Hs. Finally, the 10-m height wind speed was derived using an empirical relation between wind speed and wind wave Hs. Results showed InIRA has a good potential of retrieving wind speed.

With the rising interest in exploring the possibility of estimating geophysical parameters of interest to oceanographers by global navigation satellite system reflectometry (GNSS-R) of signals scattered from the ocean surface, and the advent of the Soil Moisture Active Passive (SMAP) [1] and the Soil Moisture and Ocean Salinity (SMOS) [2] scientific missions, there is a need to understand the relevant scattering physics at L band so as to provide a physical basis for retrieving geophysical information from ocean-scattered signals.

This paper compares the typhoon centers from the tropical cyclone best track (BT) data sets of three meteorological agencies and those from synthetic aperture radar (SAR) and infrared (IR) images. First, we carried out algorithm comparison using two newly developed algorithms and one existing wavelet-based algorithm, which were used to extract typhoon eyes in six SAR images and two IR images. These case studies showed that the extracted eyes by the three algorithms are consistent with each other. The differences among them are relatively small. However, there is a systematic difference between the extracted centers and the typhoon centers from the three BT data sets, which were interpolated to the imaging times first. We then compared the typhoon centers determined from 25 SAR and 43 IR images with those from the three BT data sets to investigate the performance of the latter at the sea surface and at the cloud top, respectively. We found that the typhoon centers from the three BT data sets are generally closer to the locations extracted from the SAR images showing sea-surface imprints of the typhoons than those from the IR images showing cloud-top structures of the typhoons.

In this study we simulate the fully polarimetric bistatic scattering behavior resulted from the interaction of electromagnetic wave with all scales of ocean waves, with a focus on the dependence of such behavior on wind direction and incidence angle. The numerical approach is our recently developed second stochastic degree iterative algorithm with sparse matrix and Chebyshev approximation. The illuminated area is 86 wavelength × 86 wavelength, which represents about 2.5 dominant wavelengths for a 3m/s Elfouhaily spectrum at L band, implying that the effect of gravity wave has been adequately included. The total number of surface unknowns is 2,130,048. Numerical results for both angular backscattering behavior and bistatic patterns are provided.

Spaceborne high-resolution synthetic aperture radar (SAR) is a potential powerful tool for rainfall pattern and intensity observations over the sea surface. However, many interesting rain-related phenomena revealed by SAR images are still not fully understood due to poor theoretical modeling of the rain–wind–wave interactions. This paper attempts to develop a physics-based radiative transfer model to capture the scattering behavior of rainfall over a rough sea surface. Raindrops are modeled as Rayleigh scattering nonspherical particles, whereas the rain-induced rough surface is described by the Log-Gaussian ring-wave spectrum. The model is validated against both empirical models and measurements. A case study of collocated Envisat ASAR data and NEXRAD rain data is presented to demonstrate the performance of the newly developed model. Finally, numerical simulation results suggest that rain-related scattering becomes significant as compared with wind-related scattering when the frequency is above C-band, whereas the raindrop volumetric scattering becomes significant above X-band.

Significant wave height (SWH) data from T/P, GFO, Jason-1 and Envisat altimeters are merged. The characteristics of SWH, such as the distributions of season average SWH, the probability distributions of different sea scales and the distributions of extreme SWH in 50 and 100 return years for China Seas and their adjacent waters are analyzed based on the merged data.

In ESA's Envisat ASAR level 2 algorithm, it is assumed that the synthetic aperture radar (SAR) image cross spectra of mixed ocean waves were the summation of SAR image cross spectra of wind waves and that of swells. But our previous studies show that in addition to this, the cross spectra of mixed waves consist of an extra term (see the companion paper submited to this symposium). Just this term leads to an inherent error of this algorithm which has not been considered yet. This paper presents an error analysis of Envisat ASAR level 2 algorithm for ocean wave spectra retrieval in different significant wave height, wavelength, wave direction and wave component conditions based on simulation technique. It shows that the inherent errors (1) change with above parameters, and are always positive which mean the retrieved ocean wave height are overestimated; (2) increase for larger significant wave height; (3) increase for smaller wavelength; (4) increases for smaller propagation angle respect to azimuth direction. (5) increase for more wind wave component. Therefore, Envisat ASAR level 2 algorithm only works in small wave height, or large wavelength, or large propagation angle, or few wind wave component conditions.

A 6-parameter frequency spectrum with two peaks and a cos-2s type spreading function is used to simulate the mixed waves. The spaceborne and airborne synthetic aperture radar (SAR) image cross spectra of mixed waves in different significant wave height, wave length, wave direction and wave component are then calculated by using Engen's nonlinear transformation formula. Analysis based on above simulation indicate that (1) the cross spectra of mixed waves dilate in range direction and shrink in azimuth direction (the so-called azimuth cutoff effect); (2) the cutoff effect increases for waves with larger wave height, or for waves with shorter wave length, or for waves propagating closer the azimuth direction, or for waves containing more wind wave component, or for spaceborne SAR; (3) the cross spectra split into two parts for waves propagating along range direction (the so-called double-peak phenomenon); (4) the direction ambiguity of ocean waves can be removed by using the imaginary part of cross spectra; (5) in addition to the contribution of wind wave part and swell part of the mixed waves, the cross spectra of mixed waves consist of an extra term which leads to an inherent error when using ESA's Envisat ASAR level 2 algorithm to retrieve ocean waves (see the companion paper submited to this symposium).

Satellite remote sensing technique has been used to monitor the oceanic environment. This paper presents the technique development and applications of satellite remote sensing to the oceanic environment in China. The technique development includes the development of algorithms and of methodology for extracting oceanographic parameters from satellite data. Applications of satellite remote sensing range from environmental monitoring to oceanographic research.

The algorithms discussed in this paper are three Constant False Alarm Rate (CFAR) models, which include the Probabilistic Neural Network(PNN) model, the K-Gamma model and the double parameters model. The SAR data utilized in the paper include ERS-2, ENVISAT and Radarsat SAR data. The data are applied in ship detection experiments and the results of ship detection of three models are compared. The results show that the PNN model's applicability is the best .The performance of PNN model in ERS and ENVISAT SAR data is better than the K-Gamma model. The K-Gamma model can only do well in Radarsat SAR data. The double parameters model can fit local distribution of SAR image in the sea.

Based on the imaging mechanics of synthetic aperture radar (SAR) mapping shallow sea bottom topography, a new method was proposed using multi-band and full-polarization SAR. If this method was used, it should need fewer in-situ data in multi-band and full-polarization SAR than single-bond and single-polarization SAR for bottom topography measurement. It should be much more accurate.

This paper presents an improved constant false alarm rate (CFAR) model for ship detection in synthetic aperture radar (SAR) imagery. The model includes the probabilistic neural networks, CFAR technique, golden section method and area growth method. It is compared with other ship detection methods. The results show that the improved CFAR model performs well

The Zhejiang-Fujian coastal front in the East China Sea has been studied using an 12-year time series (1991-2002) of Advanced Very High Resolution Radiometer (AVHRR) images. Satellite observation has shown that the front exits all year round and has a length scale of about 597 km. The strength and width of the front vary with average values of 0.14 degC/km and 14 km respectively. The front is characterized by the presence along its edges of rings which have a time scale of 1-7 days and a length of 30-90 km.

Fully polarimetric synthetic aperture radar (SAR) data collected by AIRSAR and SIR-C/X-SAR are used to analyze radar signatures of ocean features such as ocean waves, oil slicks, ship wakes, ocean fronts and submarine topography in the China Seas. The optimal polarization (OP) for the observation of these features with SAR is given based on the analyses.

It is known that coastal zone and its environments are the complex and specialized areas. Synthetic aperture radar (SAR) with all weather is the powerful tools for monitoring the dynamic changes of those regions, and fractal approaches may be a good ideal technologies for managements on the unprecedented amount of information from the coastal spatial and temporal processes and remote sensing images etc. database. The needs of coastal dynamic monitoring are introduced at the first. The imaging mechanisms and the technologies as well as the example studies of SAR detecting coastal zone are described in detail.