Waterline extraction from synthetic aperture radar (SAR) images can be transformed into a sea-land segmentation task. However, two aspects of deep learning sea-land segmentation have been ignored in the literature: 1) deep learning models are commonly built using whole images rather than focusing on their sea-land transition parts and 2) a higher resolution input may not render a better segmentation result under the constraint of a fixed-size receptive field. Our investigation on the aspects indicates that focusing the modeling process on the sea-land transition parts can benefit the waterline extraction, and the highest resolution may not be the best choice for all pixels. We proposed masked soft intersection over union (MSIoU) loss and the difference-focusing fusion decision (DFFD) ensemble learning method. MSIoU loss incorporates the mask of the transition parts to focus the modeling process on the transition parts. The DFFD ensemble learning method imitates manual labeling and can avoid selecting the resolution of input images. The DFFD ensemble model’s member models segment an image’s sea and land areas at different resolutions. Then, its fusion model further recognizes the pixels where the member models inconsistently predict sea-land types. We applied the DFFD ensemble model to 10-m-resolution SAR images of the test set. Compared to the traditional single-resolution model, the DFFD ensemble model with MSIoU loss achieved a 10.32–12.58-m accuracy in waterline extraction with a 2.08–2.78-m error reduction in the study area. Moreover, the DFFD framework is independent of data and model choice and can be readily modified for other segmentation tasks.

Satellite-based synthetic aperture radar (SAR) can provide a low-cost, frequent environment monitoring for dynamic intertidal zones. The critical problem is to realize pixel-level classification of SAR images of the intertidal zones with excellent and robust performance. Recently, deep learning, in particular deep convolutional neural networks, has provided us with promising solutions to this problem. Based on a sophisticated deep learning-based pixel-level classification model U2-Net, we propose an MB-U2-ACNet model suitable for intertidal zone land cover classification using dual-polarimetric SAR data integrated with environmental information, such as wind speed and tide level information. The MB-U2-ACNet model has a multi-branch nested U-shaped encoding-decoding structure. We extract and fuse features from multiple data sources, including satellite remote sensing and environmental information, by establishing the multi-branch structure. Furthermore, we propose an asymmetric convolution residual U-block for each encoding-decoding stage to improve the model’s feature extraction ability. Moreover, the model with attention mechanisms better distinguishes the importance of features from the channel’s perspectives and spatial dimensions. We construct a dataset with 106 Sentinel-1 SAR images from 2016 to 2020 for environment monitoring in the intertidal zone of Shanghai Nanhui. On the dataset, the proposed model reaches the overall classification accuracy of 96.40% and the mean intersection over union score of 0.8307. The experiments show the advantages of the proposed model compared with the benchmarking models due to better feature extraction and multi-source information fusion. In addition, the contributions of every added sub-structure are analyzed systematically.

In this study, a set of deep convolutional neural networks (CNNs) was designed for estimating the intensity of tropical cyclones (TCs) over the Northwest Pacific Ocean from the brightness temperature data observed by the Advanced Himawari Imager onboard the Himawari-8 geostationary satellite. We used 97 TC cases from 2015 to 2018 to train the CNN models. Several models with different inputs and parameters are designed. A comparative study showed that the selection of different infrared (IR) channels has a significant impact on the performance of the TC intensity estimate from the CNN models. Compared with the ground truth Best Track data of the maximum sustained wind speed, with a combination of four channels of data as input, the best multicategory CNN classification model has generated a fairly good accuracy (84.8%) and low root mean square error (RMSE, 5.24 m/s) and mean bias (−2.15 m/s) in TC intensity estimation. Adding attention layers after the input layer in the CNN helps to improve the model accuracy. The model is quite stable even with the influence of image noise. To reduce the side-effect of the very unbalanced distribution of TC category samples, we introduced a focal_loss function into the CNN model. After we transformed the multiclassification problem into a binary classification problem, the accuracy increased to 88.9%, and the RMSE and the mean bias are significantly reduced to 4.62 and −0.76 m/s, respectively. The results show that our CNN models are robust in estimating TC intensity from geostationary satellite images.

Synthetic aperture radar (SAR) can monitor the sea surface imprints of tropical cyclones (TCs) with high spatial resolution, day and night. Automatically locating TC center positions in SAR images is a challenging task. This article developed a two-stage, fully automatic TC-center estimation algorithm. First, the sea surface wind directions (SSWDs) at SSWD points are retrieved by the improved local gradient (ILG) method. We incrementally deflected the SSWD outward at a 0.5° angle from −50° to 10° (the negative angles represent clockwise deflection). The heat maps are generated for each of the 121 angles, and the values at each heat map are the cumulative numbers of the lines perpendicular to the compensated SSWDs. The site corresponding to the maximum cumulative number in all 121 heat maps is the coarsely estimated center position. This center search is the culmination if it falls outside the SAR image. Otherwise, the second stage is triggered, and the sub-SAR image (150 km$\times150$km) centered at the coarsely estimated center position is extracted. Then, the first-stage procedure is repeated with the sub-SAR image to precisely estimate the center position. Optionally, the precisely estimated center position can be further adjusted by considering that normalized radar cross section (NRCS) is normally minimal at the TC center. We applied the algorithm to 87 SAR images. Five of these images do not contain TC centers. The results are in good agreement with the visually located TC center positions and those in the best track (BT) datasets.

Recently, deep learning has already shown its availability in synthetic aperture radar (SAR) image classification. To improve the deep learning model's performance on multisource remote sensing information fusion, we propose a multi-branch deep convolutional neural network model specially tailored from the U2-Net framework and with asymmetric convolutions. We name it the MB-U2-ACNet model. Based on experiments on a constructed dataset dedicated for Shanghai Nanhui Dongtan intertidal zone environment monitoring, it is verified that the proposed MB-U2-ACNet model has better performance than the existing representative deep and traditional methods.

Researchers have already verified that the deep learning (DL) technology can realize accurate and robust mapping of tropical cyclone-induced coastal inundation in synthetic aperture radar imagery. In order to liberate the DL-based inundation mapping from human supervision, we propose to use the clustering of deep convolutional autoencoder-generated features. The mapping results of Lekima 2019-induced inundation demonstrate the advantages and availability of the proposed method.

Accurate and robust classification methods of sea ice and open water are significant for many applications. Synthetic Aperture Radar (SAR) imaging capability is independent of weather conditions and is widely used in sea ice classification. U-Net, a deep learning framework, has achieved great success in the field of biomedical image classification. In this study, we construct a U-Net-based “end-to-end” model to classify the sea ice and open water pixels in SAR imagery. Five SAR images acquired in the Gulf of Alaska near Bering Strait are used in this case study. We manually label the SAR images as ice and water. The labeled images from the first four SAR image are divided into chips to be fed into the U-Net model for training. The fifth SAR image is employed as the testing data. Experiments show that the precision and the recall of the testing image is 91.64% and 91.70%, respectively. Most of the sea ice, including small chunks and sinuous ice edges, can be successfully classified.

In this study, we proposed an automatic internal wave (IW) signature extraction method based on the deep convolutional neural networks (DCNN). Our objective is to provide a rapid and simple to use method that can tackle the IW signature extraction in images from different satellite sensors without re-training or manual interference. We proved the generalization ability of our method across multiple optical satellite sensors. The statistical results show this DCNN-based method has appreciable transferability and is promising for efficient extraction of internal wave signature in different satellite images with varying spatial resolution even under complex imaging conditions.

The sea surface wind at low and moderate speed (20 m/s). In this study, we explore to use a novel deep convolutional neural networks (DCNN) architecture, U-Net, to retrieve the sea surface wind at high speed. The Sentinel-1B SAR cross-polarized (HV and VH) Normalized Radar Cross Section (NRCS), SAR incidence angle and the ancillary Global Forecast System (GFS) model wind directions are used as the U-Net input data. The GFS model wind speed is the ground-truth data. The results suggest that the U-Net wind retrieval model has a capability to retrieve the high wind speed from SAR data with the physical model-based data. The accuracy of physical model data directly affects the U-Net model results. Meanwhile, U-Net model has a better continuity in the joint area of two strips.

In the past few years, artificial intelligent (AI) technology has been widely used in many research fields for big data information mining and shown great potential applications in computer vision, natural language processing, bioinformatics, among others. In the area of remote sensing oceanography, we categorize its applications in four major categories: image classification, data fusion, algorithm development and oceanic phenomenon forecast. In this paper, we present two examples to demonstrate such applications. In the first example we applied a well-studied AI framework, U-Net, to a NASA JPL’s UAVSAR Synthetic Aperture Radar (SAR) image to classify coastal zone types in the Gulf Coast of USA. In the second example, we trained the AI framework, LSTM model, using the time series of blended microwave Sea Surface Temperature (SST) data, and made the equatorial SST pattern forecast. Validation studies in both cases showed the robustness of AI-based technology for oceanography research.

The marine atmospheric boundary layer (MABL) roll plays an important role in the turbulent exchange of momentum, sensible heat, and moisture throughout the boundary layer of tropical cyclones. Hence, rolls are believed to be closely related to storm development and intensification. In this study, based on the RADARSAT-2 dataset including various tropical cyclone (TC) intensities, the roll characteristics are retrieved from synthetic aperture radar (SAR) images via fast Fourier transform (FFT). We investigate the roll wavelengths at variance of TC intensities and found that the roll wavelengths are related to the distance with respect to TC center and TC intensities. This study is promising to bring roll-induced effects into hurricane forecasting model.

In this paper, we present a method of ship size extraction for Sentinel-1 synthetic aperture radar (SAR) images, which is composed of the image processing stage and the regression stage. In order to achieve extraction with high accuracy, considering the data characteristics of Sentinel-1 images, we propose to use the dual-polarization fusion and the nonlinear regression with the gradient boosting. The experiments and analyses on a relatively large data set show that: 1) compared with the existing and related studies, the proposed method achieves an improved performance. The extraction errors are pushed under one pixel, and they are 4.66% (8.80 m) and 7.01% (2.17 m) for length and width, respectively; 2) the dual-polarization information fusion does improve the size extraction accuracy; and 3) the nonlinear regression does exploit the relationship between the influential factors and the size parameters and provide a better performance than the linear regression. The experimental results verify that the proposed design is suitable for ship size extraction in Sentinel-1 SAR images.

In order that global, dominant, and complete outlines of land covers are delineated, in this paper, we propose a regularized L0gradient minimization method which is specially developed for segmenting polarimetric synthetic aperture radar (PolSAR) images, and present a region-based stepwise design to implement it. The performance of the proposed method is tested and analyzed on two experimental data sets, with visual presentation as well as numerical evaluation. They both confirm that the proposed method achieves its principal goal and demonstrate its availability and advantage as a pre-processing step for PolSAR image interpretation chains.

It is believed that it is essential to take the spatial adaptivity into the segmentation method for polarimetric synthetic aperture radar (PolSAR) images. The size and shape of each segment and the strength of the relationship of neighboring pixels need to depend on the local spatial complexity of the scene. The wedgelet framework provides a promising analysis tool for spatial information. The major advantage of the wedgelet analysis is that it captures the geometrical structure of images at multiple scales, with the local spatial complexity taken into consideration. Hence, in this paper, we propose a wedgelet approximation and analysis framework specially designed for PolSAR data. Based on this framework, a spatially adaptive representation and segmentation method is constructed and presented. It mainly consists of three parts: first, the multiscale wedgelet decomposition is applied to the PolSAR image, and the local geometrical information is captured in an optimal way; then, the image is segmented in a spatially adaptive manner by the multiscale wedgelet representation in the form of the regularized optimization, which keeps a balance between the approximation and parsimony of the representation; the final part is the spatial-complexity-adaptive segmentation refinement based on the Wishart Markov random field model. The performance of the proposed method is presented and analyzed on two experimental data sets, with visual presentation and numerical evaluation. It is also compared with an existing and theoretically well-founded segmentation method. The experiments and results demonstrate the availability and advantage of the proposed method.

In this paper, a multi-temporal (MT) superpixel generation method is proposed. MT superpixels are generated based on a novel edge extraction method designed for high resolution (HR) synthetic aperture radar (SAR) images and using MT spatial information in an optimization way. Numerical evaluation and comparison on simulated and real data sets demonstrate its availability for MT HR SAR image analysis.

The classic region-based filter for edge extraction for polarimetric synthetic aperture radar images is theoretically founded and efficient. However, in practical use, its performance is limited because the assumption of independence and identical distribution is often not met, particularly in heterogeneous areas. In this letter, we present a degenerate filter design integrated with the weighted maximum likelihood estimation to overcome this limitation. The performance of the proposed methodology is presented and analyzed on both simulated and real experimental data sets using visual presentation, as well as numerical evaluation and comparison with the classic method. They both demonstrate the availability and advantage of the proposed method.

In this paper, we present an image processing chain that can interpret high resolution synthetic aperture radar (SAR) imagery for building layover characterization and extraction in urban areas. It is composed of three main parts - generation of hint areas, generation of superpixels, and optimized cut of layovers via superpixel merging. The proposed framework is complete, and flexibly integrates necessary information, both area and boundary, for building layover extraction; the experimental results show that its performance is promising.

Ship detection is an important application in remote sensing and Earth observation since last century. Properties of ship targets changed greatly as observation accuracy of SAR sensors is strongly increased. Ship targets become objects detailed in structure by contrast to point targets in lower resolution SAR images. In this paper, we present a radiometric-spatial analysis (RSA) based method for ship detection in high resolution SAR images.

Polarimetric synthetic aperture radar (PolSAR) image classification, an important technique in the remote sensing area, has been deeply studied for a couple of decades. In order to develop a robust automatic or semiautomatic classification system for PolSAR images, two important problems should be addressed: 1) incorporation of spatial relations between pixels; 2) estimation of the number of classes in the image. Therefore, in this paper, we present a novel superpixel-based classification framework with an adaptive number of classes for PolSAR images. The approach is mainly composed of three operations. First, the PolSAR image is partitioned into superpixels, which are local, coherent regions and preserve most of the characteristics necessary for image information extraction. Then, the number of classes and each class center within the data are estimated using the pairwise dissimilarity information between superpixels, followed by the final classification operation. The proposed framework takes the spatial relations between pixels into consideration and makes good use of the inherent statistical characteristics and contour information of PolSAR data. The framework is capable of improving the classification accuracy, making the results more understandable and easier for further analyses, and providing robust performance under various numbers of classes. The performance of the proposed classification framework on one synthetic and three real data sets is presented and analyzed; and the experimental results show that the framework provides a promising solution for unsupervised classification of PolSAR images.

In this paper, we improve the traditional bag-of-words-based image representation method in two aspects: preserving the semantics in vocabulary generation and incorporating spatial relations in image representation. Based on that, we present a novel context-aware information modeling method for high resolution synthetic aperture radar image scene interpretation. We compare the proposed method with traditional ones in scene interpretation on TerraSAR-X data sets.

In this paper, we propose a coarse-to-fine framework design and implementation for oil tank detection in optical satellite imagery. The framework is mainly composed of two operations: 1) from the whole scene imagery, extraction of patches with oil tanks based on the probabilistic latent semantic analysis model; 2) in the relatively small size patches, detection of the oil tanks with Hough transform and template matching. Experiments show that the framework provides a promising solution for oil tank detection in optical satellite imagery.

In this paper, we present a number-of-classes-adaptive unsupervised classification framework for synthetic aperture radar (SAR) images. The framework aims at the provision of robust classification for SAR images even if the number of classes existing in the scene is unknown. It mainly consists of estimation of the number of classes, extraction of each class center, classification of image patches, and integration of spatial relations between patches. The experiment on a TerraSAR-X SAR image shows that the proposed framework presents a promising performance for SAR image classification.

In this paper, we present a scene interpretation framework for Synthetic Aperture Radar (SAR) images, using keywords of the image contents provided by users. The framework consists of incorporation of prior knowledge with SAR iMage Annotation Tool (SARMAT), representation of SAR images, and prediction of scene labels based on the supervised Latent Dirichlet Allocation (sLDA) model. The experiment on a TerraSAR-X SAR image shows that the proposed framework provides a promising performance for SAR image scene interpretation.

In this letter, we present a novel foreground/background separation (FBS) framework for interpreting polarimetric synthetic aperture radar (PolSAR) images. The FBS framework takes the spatial relations between pixels into consideration and incorporates the advantages of pairwise dissimilarity-based grouping schemes. The FBS method can separate specific targets and objects from the background, which is essential in an interpretation system. Multiple FBS operations can be integrated to interpret PolSAR images, flexibly fusing various inherent features of PolSAR data. Several PolSAR data sets are used to verify the proposed approach.