Search

Synthetic aperture radar (SAR) is often affected by other high-power electromagnetic devices during ground observation, which causes unintentional radio frequency interference (RFI) with the acquired echo, bringing adverse effects into data processing and image interpretation. When faced with the task of screening massive SAR data, there is an urgent need for the global perception and detection of interference. The existing RFI detection method usually only uses a single type of data for detection, ignoring the information association between the data at all levels of the real SAR product, resulting in some computational redundancy. Meanwhile, current deep learning-based algorithms are often unable to locate the range of RFI coverage in the azimuth direction. Therefore, a novel RFI processing framework from quick-looks to single-look complex (SLC) data and then to raw echo is proposed. We take the data of Sentinel-1 terrain observation with progressive scan (TOPS) mode as an example. By combining the statistics-assisted network with the sliding-window algorithm and the error-tolerant training strategy, it is possible to accurately detect and locate RFI in the quick looks of an SLC product. Then, through the analysis of the TOPSAR imaging principle, the position of the RFI in the SLC image is preliminarily confirmed. The possible distribution of the RFI in the corresponding raw echo is further inferred, which is one of the first attempts to use spaceborne SAR data to elucidate the RFI location mapping relationship between image data and raw echo. Compared with directly detecting all of the SLC data, the time for the proposed framework to determine the RFI distribution in the SLC data can be shortened by 53.526%. All the research in this paper is conducted on Sentinel-1 real data, which verify the feasibility and effectiveness of the proposed framework for radio frequency signals monitoring in advanced spaceborne SAR systems.

Graph neural networks (GNNs) are effective for structured data analysis but face reduced learning accuracy due to noisy connections and the necessity for explicit graph structures and labels. This requirement constrains their usability in diverse graph-based applications. In order to address these issues, considerable research has been directed toward graph structure learning that aims to denoise graph structures concurrently and refine GNN parameters. However, existing graph structure learning approaches encounter several challenges, including dependence on label information, underperformance of learning algorithms, insufficient data augmentation methods, and limitations in performing downstream tasks. We propose Uogtag, an unsupervised graph structure learning framework to address these challenges. Uogtag optimizes graph topology through the selection of suitable graph learners for the input data and incorporates contrastive learning with adaptive data augmentation, enhancing the learning and applicability of graph structures for downstream tasks. Comprehensive experiments on various real-world datasets demonstrate Uogtag’s efficacy in managing noisy graphs and label scarcity.

Cyber-physical-social systems (CPSS) have recently gained attention from researchers due to their combination of cyber, physical, and social spaces. Modeling and Analysis of Real-Time and Embedded systems (MARTE) is a Unified Modeling Language (UML) extension profile that supports the specification, design, and verification of Real-Time Embedded Systems (RTES). While MARTE Statecharts can assist in describing CPS, it does not model the uncertainty within a CPSS environment. To enhance the accuracy of CPSS analysis, we propose the stohMCharts (stochastic hybrid MARTE statecharts) modelling framework as an extension of MARTE statecharts for modelling and analyzing stochastic hybrid systems. stohMCharts can model CPSS in a unified manner. Additionally, based on the mapping rules and algorithms, we have developed a tool to convert models built in stohMChart language into Networks Stochastic Hybrid Automata (NSHA) which can be verified by statistical model checker UPPAAL-SMC. We demonstrate the efficiency and accuracy of the framework by applying it to one autonomous driving scenarios.

Ship recognition with optical remote sensing images is currently widely used in fishery management, ship traffic surveillance, and maritime warfare. However, it currently faces two major challenges: recognizing rotated targets and achieving fine-grained recognition. To address these challenges, this paper presents a new model called Related-YOLO. This model utilizes the mechanisms of relational attention to stress positional relationships between the components of a ship, extracting key features more accurately. Furthermore, it introduces a hierarchical clustering algorithm to implement adaptive anchor boxes. To tackle the issue of detecting multiple targets at different scales, a small target detection head is added. Additionally, the model employs deformable convolution to extract the features of targets with diverse shapes. To evaluate the performance of the proposed model, a new dataset named FGWC-18 is established, specifically designed for fine-grained warship recognition. Experimental results demonstrate the excellent performance of the model on this dataset and two other public datasets, namely FGSC-23 and FGSCR-42. In summary, our model offers a new route to solve the challenging issues of detecting rotating targets and fine-grained recognition with remote sensing images, which provides a reliable foundation for the application of remote sensing images in a wide range of fields.

Recently, deep learning methods have been widely studied in the field of polarimetric synthetic aperture radar (PolSAR) ship detection. However, extracting polarimetric and spatial features on the whole PolSAR image will result in high computational complexity. In addition, in the massive data ship detection task, the image to be detected contains a large number of invalid areas, such as land and seawater without ships. Therefore, using ship coarse detection methods to quickly locate the potential areas of ships, that is, ship potential area extraction, is an important prerequisite for PolSAR ship detection. Since existing unsupervised PolSAR ship detection methods based on pixel-level features often rely on fine sea–land segmentation pre-processing and have poor applicability to images with complex backgrounds, in order to solve the abovementioned issue, this paper proposes a PolSAR ship potential area extraction method based on the neighborhood semantic differences of an LDA bag-of-words topic model. Specifically, a polarimetric feature suitable for the scattering diversity condition is selected, and a polarimetric feature map is constructed; the superpixel segmentation method is used to generate the bag of words on the feature map, and latent high-level semantic features are extracted and classified with the improved LDA bag-of-words topic model method to obtain the PolSAR ship potential area extraction result, i.e., the PolSAR ship coarse detection result. The experimental results on the self-established PolSAR dataset validate the effectiveness and demonstrate the superiority of our method.

Object detection under weakly supervised learning is a challenging issue. In a remote sensing ship detection task, the weakly supervised learning method requires that the training set only has image-level class annotations. In the absence of location information, it is difficult to locate ships and extract features. Moreover, when the detector extracts more than one candidate region, the image-level annotation makes it difficult to determine whether the candidate regions are multiple ships or mixed with the detected background. To address these issues, this article analyzes the interaction between class information and location information, and proposes a weakly supervised detection method, i.e., PistonNet, based on data-efficient image transformers. PistonNet proposes an artificial point, which is inserted into the feature map. Artificial point can suppress the background and enhance the object by interfering in the weight distribution of object and background in self-attention calculation. PistonNet also proposes joint confidence probability to improve detection accuracy. Experiments on the GF1-LRSD and NWPU VHR-10 datasets show that the proposed method boosts detection performance effectively. The main improvements of PistonNet as well as the contributions of this article are threefold. First, PistonNet is a specific weakly supervised object detection method for single-class detection, which provides an innovative approach to classifying target and background. Second, PistonNet reaches the level of advanced supervised detectors on detection accuracy with fewer parameters. Finally, the objects’ locations detected by PistonNet are obtained by segmenting regions on the heat map. PistonNet’s background suppression ability makes it free from dependence on segmentation threshold.

Detecting infrared (IR) small moving targets in complex scenes quickly, accurately, and robustly remains a challenging problem in the current research field. To address this issue, this paper proposes a novel spatial–temporal block-matching patch-tensor (STBMPT) model based on a low-rank sparse decomposition (LRSD) framework. This model enhances the traditional infrared patch-tensor (IPT) model by incorporating joint spatial–temporal sampling to exploit inter-frame information and constructing a low-rank patch tensor using image block matching. Furthermore, a novel prior-weight calculation is introduced, utilizing the eigenvalues of the local structure tensor to suppress interference such as strong edges, corners, and point-like noise from the background. To improve detection efficiency, the tensor is constructed using a matching group instead of using a traditional sliding window. Finally, the background and target components are separated using the alternating direction method of multipliers (ADMM). Qualitative and quantitative experimental analysis in various scenes demonstrates the superior detection performance and efficiency of the proposed algorithm for detecting infrared small moving targets in complex scenes.

Despite significant advancements in remote sensing object tracking (RSOT) in recent years, achieving accurate and continuous tracking of tiny-sized targets remains a challenging task due to similar object interference and other related issues. In this paper, from the perspective of feature enhancement and a better feature matching strategy, we present a tracker SiamTM specifically designed for RSOT, which is mainly based on a new target information enhancement (TIE) module and a multi-level matching strategy. First, we propose a TIE module to address the challenge of tiny object sizes in satellite videos. The proposed TIE module goes along two spatial directions to capture orientation and position-aware information, respectively, while capturing inter-channel information at the global 2D image level. The TIE module enables the network to extract discriminative features of the targets more effectively from satellite images. Furthermore, we introduce a multi-level matching (MM) module that is better suited for satellite video targets. The MM module firstly embeds the target feature map after ROI Align into each position of the search region feature map to obtain a preliminary response map. Subsequently, the preliminary response map and the template region feature map are subjected to the Depth-wise Cross Correlation operation to get a more refined response map. Through this coarse-to-fine approach, the tracker obtains a response map with a more accurate position, which lays a good foundation for the prediction operation of the subsequent sub-networks. We conducted extensive experiments on two large satellite video single-object tracking datasets: SatSOT and SV248S. Without bells and whistles, the proposed tracker SiamTM achieved competitive results on both datasets while running at real-time speed.

Synthetic aperture radar (SAR) automatic target recognition (ATR) has been widely applied in multiple fields. However, the special imaging mechanism of SAR results in different visual features of the same target at different azimuth angles, so single-aspect SAR target recognition has the limitation of observing the target from a single perspective. Multi-aspect SAR target recognition technology can overcome this limitation by utilizing information from different azimuths and effectively improve target recognition performance. Considering the order dependency and data limitation of existing methods, this paper proposes a multi-aspect SAR recognition method based on Non-Local, which applies a self-attention calculation to feature maps to learn the correlation between multi-aspect SAR images. Meanwhile, in order to improve the generalization ability of the proposed method under limited data, a network based on contrastive learning was designed to pre-train the feature extraction part of the whole network. The experimental results using the MSTAR dataset show that the proposed method has excellent recognition accuracy and good robustness.

False targets caused by multichannel Synthetic Aperture Radar (SAR) are similar to a defocused ship in both shape and texture, making it difficult to discriminate in the full-aperture SAR image. To address the issue of false alarms caused by such false targets, this paper proposes a multichannel SAR false-target discrimination method based on sub-aperture and full-aperture feature learning. First, amplitude calculation is performed on complex SAR images to obtain the amplitude images, and transfer learning is utilized to extract the full-aperture features from the amplitude images. Then, sub-aperture decomposition is performed on complex SAR images to obtain a series of sub-aperture images, and the Stacked Convolutional Auto-Encoders (SCAE) are applied to extract the sub-aperture features from the sub-aperture images. Finally, the sub-aperture and the full-aperture features are concatenated to form the joint features, which are used to accomplish target discrimination. The accuracy of the method proposed in this paper is 16.32% higher than that of the approach only using the full-aperture feature on GF-3 UFS SAR images.

Deep convolutional neural networks (DCNNs) have attracted extensive attention in synthetic aperture radar (SAR) image target detection. However, there is usually a lack of labeled data in SAR images which results in difficulties to improve the detection performance with DCNN based methods. Transfer learning borrows generic knowledge from other tasks for the target task and can be an effective solution to the small sample problem, but the features' generality and specificity across various models are not discussed in target detection. In addition, conventional DCNN based models have too many parameters related to the prior information of the targets which increase the difficulty of transfer learning between different tasks. In this paper, we mainly conduct research in three folds to address these mentioned issues. First, the generality and specificity of features in different network models are analyzed in the field of SAR target detection. Second, a transitive transfer framework is developed for detection under small sample condition. Third, an anchor free rotatable detector with flexible structure is designed for SAR images detection to suit the transfer detection framework proposed in this paper and a reasonable explanation is given for its upper hand with small samples. Experimental results demonstrate that: 1) transitive transfer framework is valid in SAR target detection; 2) the anchor free detector in this framework is superior to the anchor based one; 3) the proposed detector achieves better performance under small sample condition by comparison with state-of-the-art methods.

Deep learning has been successfully utilized in synthetic aperture radar (SAR) automatic target recognition tasks and obtained state-of-the-art results. However, current deep learning algorithms do not perform well when SAR images are occluded, noisy, or with a great depression angle variance. This article proposes a novel method, SAR capsule network, to achieve the accurate and robust classification of SAR images without significantly increasing network complexity. Specifically, we develop a convolutional neural network extension based on Hinton's capsule network to capture spatial relationships specialized in classification between different entities in a SAR image. The SAR capsules are learned by a vector-based full connection operation instead of the traditional routing process, which not only alleviates the computational burden but also improves recognition accuracy. For occlusion, additive noise, and multiplicative noise tests, SAR capsule network shows superior robustness compared with typical convolution neural networks. When missing training data in a certain aspect angle range or existing a large depression angle variance between training data and test data, the proposed network achieves better performance than the existing works and reveals some competitive advantages in several test scenarios.

A multi-modal 3D object-detection method, based on data from cameras and LiDAR, has become a subject of research interest. PointPainting proposes a method for improving point-cloud-based 3D object detectors using semantic information from RGB images. However, this method still needs to improve on the following two complications: first, there are faulty parts in the image semantic segmentation results, leading to false detections. Second, the commonly used anchor assigner only considers the intersection over union (IoU) between the anchors and ground truth boxes, meaning that some anchors contain few target LiDAR points assigned as positive anchors. In this paper, three improvements are suggested to address these complications. Specifically, a novel weighting strategy is proposed for each anchor in the classification loss. This enables the detector to pay more attention to anchors containing inaccurate semantic information. Then, SegIoU, which incorporates semantic information, instead of IoU, is proposed for the anchor assignment. SegIoU measures the similarity of the semantic information between each anchor and ground truth box, avoiding the defective anchor assignments mentioned above. In addition, a dual-attention module is introduced to enhance the voxelized point cloud. The experiments demonstrate that the proposed modules obtained significant improvements in various methods, consisting of single-stage PointPillars, two-stage SECOND-IoU, anchor-base SECOND, and an anchor-free CenterPoint on the KITTI dataset.

Deep learning methods have been widely studied in the field of polarimetric synthetic aperture radar (PolSAR) ship detection over the past few years. However, the backscattering of manmade targets, including ships, is sensitive to the relative geometry between target orientation and radar line of sight, which makes the diversity of polarimetric and spatial features of ships. The diversity of scattering leads to a relative increase in the scarcity of PolSAR-labeled samples, which are difficult to obtain. To solve the abovementioned issue and extract the polarimetric and spatial features of PolSAR images better, this paper proposes a few-shot PolSAR ship detection method based on the combination of constructed polarimetric input data selection and improved contrastive self-supervised learning (CSSL) pre-training. Specifically, eight polarimetric feature extraction methods are adopted to construct deep learning network input data with polarimetric features. The backbone is pre-trained with un-labeled PolSAR input data through an improved CSSL method without negative samples, which enhances the representation capability by the multi-scale feature fusion module (MFFM) and implements a regularization strategy by the mix-up auxiliary pathway (MUAP). The pre-trained backbone is applied to the downstream ship detection network; only a few labeled samples are used for fine-tuning and the construction method of polarimetric input data with the best detection effect is studied. The comparison and ablation experiment results on the self-established PolSAR ship detection dataset verify the superiority of the proposed method, especially in the case of few-shot learning.

In the combat system, infrared target detection is an important issue worthy of study. However, due to the small size of the target in the infrared image, the low signal-to-noise ratio of the image and the uncertainty of motion, how to detect the target accurately and quickly is still difficult. Therefore, in this paper, an infrared method of detecting small moving targets based on a coarse-to-fine structure (MCFS) is proposed. The algorithm mainly consists of three modules. The potential target extraction module first smoothes the image through a Laplacian filter and extracts the prior weight of the image by the proposed weighted harmonic method to enhance the target and suppress the background. Then, the local variance feature map and local contrast feature map of the image are calculated through a multiscale three-layer window to obtain the potential target region. Next, a new robust region intensity level (RRIL) algorithm is proposed in the spatial-domain weighting module. Finally, the temporal-domain weighting module is established to enhance the target positions by analyzing the kurtosis features of temporal signals. Experiments are conducted on real infrared datasets. Through scientific analysis, the proposed method can successfully detect the target, at the same time, the ability to suppress the background and the ability to improve the target has reached the maximum, which verifies the effectiveness of the algorithm.

Ship detection from synthetic aperture radar (SAR) images has become a major research field in recent years. It plays a major role in monitoring the ocean, marine rescue activities, and marine safety warnings. However, there are still some factors that restrict further improvements in detecting performance, e.g., multi-scale ship transformation and unfocused images caused by motion. In order to resolve these issues, in this paper, a doppler feature matrix fused with a multi-layer feature pyramid network (D-MFPN) is proposed for SAR ship detection. The D-MFPN takes single-look complex image data as input and consists of two branches: the image branch designs a multi-layer feature pyramid network to enhance the positioning capacity for large ships combined with an attention module to refine the feature map’s expressiveness, and the doppler branch aims to build a feature matrix that characterizes the ship’s motion state by estimating the doppler center frequency and frequency modulation rate offset. To confirm the validity of each branch, individual ablation experiments are conducted. The experimental results on the Gaofen-3 satellite ship dataset illustrate the D-MFPN’s optimal performance in defocused ship detection tasks compared with six other competitive convolutional neural network (CNN)-based SAR ship detectors. Its satisfactory results demonstrate the application value of the deep-learning model fused with doppler features in the field of SAR ship detection.

Optical remote-sensing images have a wide range of applications, but they are often obscured by clouds, which affects subsequent analysis. Therefore, cloud removal becomes a necessary preprocessing step. In this paper, a novel and superior transformer-based network is proposed, named Cloudformer. The proposed method novelly combines the advantages of convolution and a self-attention mechanism: it uses convolution layers to extract simple features over a small range in the shallow layer, and exerts the advantage of a self-attention mechanism in extracting correlation in a large range in the deep layer. This method also introduces Locally-enhanced Positional Encoding (LePE) to flexibly generate suitable positional encodings for different inputs and to utilize local information to enhance encoding capabilities. Exhaustive experiments on public datasets demonstrate the superior ability of the method to remove both thin and thick clouds, and the effectiveness of the proposed modules is validated by ablation studies.

Continuous frames of point-cloud-based object detection is a new research direction. Currently, most research studies fuse multi-frame point clouds using concatenation-based methods. The method aligns different frames by using information on GPS, IMU, etc. However, this fusion method can only align static objects and not moving objects. In this paper, we proposed a non-local-based multi-scale feature fusion method, which can handle both moving and static objects without GPS- and IMU-based registrations. Considering that non-local methods are resource-consuming, we proposed a novel simplified non-local block based on the sparsity of the point cloud. By filtering out empty units, memory consumption decreased by 99.93%. In addition, triple attention is adopted to enhance the key information on the object and suppresses background noise, further benefiting non-local-based feature fusion methods. Finally, we verify the method based on PointPillars and CenterPoint. Experimental results show that the mAP of the proposed method improved by 3.9% and 4.1% in mAP compared with concatenation-based fusion modules, PointPillars-2 and CenterPoint-2, respectively. In addition, the proposed network outperforms powerful 3D-VID by 1.2% in mAP.

In recent years, synthetic aperture radar (SAR) automatic target recognition (ATR) has been widely used in both military and civilian fields. Due to the sensitivity of SAR images to the observation azimuth, the multi-aspect SAR image sequence contains more information for recognition than a single-aspect one. Nowadays, multi-aspect SAR target recognition methods mainly use recurrent neural networks (RNN), which rely on the order between images and thus suffer from information loss. At the same time, the training of the deep learning model also requires a lot of training data, but multi-aspect SAR images are expensive to obtain. Therefore, this paper proposes a multi-aspect SAR recognition method based on self-attention, which is used to find the correlation between the semantic information of images. Simultaneously, in order to improve the anti-noise ability of the proposed method and reduce the dependence on a large amount of data, the convolutional autoencoder (CAE) used to pretrain the feature extraction part of the method is designed. The experimental results using the MSTAR dataset show that the proposed multi-aspect SAR target recognition method is superior in various working conditions, performs well with few samples and also has a strong ability of anti-noise.

Infrared imaging plays an important role in space-based early warning and anti-missile guidance due to its particular imaging mechanism. However, the signal-to-noise ratio of the infrared image is usually low and the target is moving, which makes most of the existing methods perform inferiorly, especially in very complex scenes. To solve these difficulties, this paper proposes a novel multi-frame spatial–temporal patch-tensor (MFSTPT) model for infrared dim and small target detection from complex scenes. First, the method of simultaneous sampling in spatial and temporal domains is adopted to make full use of the information between multi-frame images, establishing an image-patch tensor model that makes the complex background more in line with the low-rank assumption. Secondly, we propose utilizing the Laplace method to approximate the rank of the tensor, which is more accurate. Third, to suppress strong interference and sparse noise, a prior weighted saliency map is established through a weighted local structure tensor, and different weights are assigned to the target and background. Using an alternating direction method of multipliers (ADMM) to solve the model, we can accurately separate the background and target components and acquire the detection results. Through qualitative and quantitative analysis, experimental results of multiple real sequences verify the rationality and effectiveness of the proposed algorithm.

With the development of imaging and space-borne satellite technology, a growing number of multipolarized SAR imageries have been implemented for object detection. However, most of the existing public SAR ship datasets are grayscale images under single polarization mode. To make full use of the polarization characteristics of multipolarized SAR, a dual-polarimetric SAR dataset specifically used for ship detection is presented in this paper (DSSDD). For construction, 50 dual-polarimetric Sentinel-1 SAR images were cropped into 1236 image slices with the size of 256 × 256 pixels. The variances and covariance of both VV and VH polarization were fused into R,G,B channels of the pseudo-color image. Each ship was labeled with both a rotatable bounding box (RBox) and a horizontal bounding box (BBox). Apart from 8-bit pseudo-color images, DSSDD also provides 16-bit complex data for readers. Two prevalent object detectors R3Det and Yolo-v4 were implemented on DSSDD to establish the baselines of the detectors with the RBox and BBox respectively. Furthermore, we proposed a weakly supervised ship detection method based on anomaly detection via advanced memory-augmented autoencoder (MemAE), which can significantly remove false alarms generated by the two-parameter CFAR algorithm applied upon our dual-polarimetric dataset. The proposed advanced MemAE method has the advantages of a lower annotation workload, high efficiency, good performance even compared with supervised methods, making it a promising direction for ship detection in dual-polarimetric SAR images. The dataset is available on github.

Recently, deep learning-based methods have made great improvements in object detection in remote sensing images (RSIs). However, detecting tiny objects in low-resolution images is still challenging. The features of these objects are not distinguishable enough due to their tiny size and confusing backgrounds and can be easily lost as the network deepens or downsamples. To address these issues, we propose an effective Tiny Ship Detector for Low-Resolution RSIs, abbreviated as LR-TSDet, consisting of three key components: a filtered feature aggregation (FFA) module, a hierarchical-atrous spatial pyramid (HASP) module, and an IoU-Joint loss. The FFA module captures long-range dependencies by calculating the similarity matrix so as to strengthen the responses of instances. The HASP module obtains deep semantic information while maintaining the resolution of feature maps by aggregating four parallel hierarchical-atrous convolution blocks of different dilation rates. The IoU-Joint loss is proposed to alleviate the inconsistency between classification and regression tasks, and guides the network to focus on samples that have both high localization accuracy and high confidence. Furthermore, we introduce a new dataset called GF1-LRSD collected from the Gaofen–1 satellite for tiny ship detection in low-resolution RSIs. The resolution of images is 16m and the mean size of objects is about 10.9 pixels, which are much smaller than public RSI datasets. Extensive experiments on GF1-LRSD and DOTA-Ship show that our method outperforms several competitors, proving its effectiveness and generality.

Ship detection is a significant and challenging task in remote sensing. At present, due to the faster speed and higher accuracy, the deep learning method has been widely applied in the field of ship detection. In ship detection, targets usually have the characteristics of arbitrary-oriented property and large aspect ratio. In order to take full advantage of these features to improve speed and accuracy on the base of deep learning methods, this article proposes an anchor-free method, which is referred as CPS-Det, on ship detection using rotatable bounding box. The main improvements of CPS-Det as well as the contributions of this article are as follows. First, an anchor-free based deep learning network was used to improve speed with fewer parameters. Second, an annotation method of oblique rectangular frame is proposed, which solves the problem that periodic angle and bounded coordinates in conjunction with the regression calculation can lead to the problem of loss anomalies. For the annotation scheme proposed in this paper, a scheme for calculating Angle Loss is proposed, which makes the loss function of angle near the boundary value more accurate and greatly improves the accuracy of angle prediction. Third, the centerness calculation of feature points is optimized in this article so that the center weight distribution of each point is suitable for the rotation detection. Finally, a scheme combining centerness and positive sample screening is proposed and its effectiveness in ship detection is proved. Experiments on remote sensing public dataset HRSC2016 show the effectiveness of our approach.

At present, synthetic aperture radar (SAR) automatic target recognition (ATR) has been deeply researched and widely used in military and civilian fields. SAR images are very sensitive to the azimuth aspect of the imaging geomety; the same target at different aspects differs greatly. Thus, the multi-aspect SAR image sequence contains more information for classification and recognition, which requires the reliable and robust multi-aspect target recognition method. Nowadays, SAR target recognition methods are mostly based on deep learning. However, the SAR dataset is usually expensive to obtain, especially for a certain target. It is difficult to obtain enough samples for deep learning model training. This paper proposes a multi-aspect SAR target recognition method based on a prototypical network. Furthermore, methods such as multi-task learning and multi-level feature fusion are also introduced to enhance the recognition accuracy under the case of a small number of training samples. The experiments by using the MSTAR dataset have proven that the recognition accuracy of our method can be close to the accruacy level by all samples and our method can be applied to other feather extraction models to deal with small sample learning problems.

Clouds are one of the most serious disturbances when using satellite imagery for ground observations. The semi-translucent nature of thin clouds provides the possibility of 2D ground scene reconstruction based on a single satellite image. In this paper, we propose an effective framework for thin cloud removal involving two aspects: a network architecture and a training strategy. For the network architecture, a Wasserstein generative adversarial network (WGAN) in YUV color space called YUV-GAN is proposed. Unlike most existing approaches in RGB color space, our method performs end-to-end thin cloud removal by learning luminance and chroma components independently, which is efficient at reducing the number of unrecoverable bright and dark pixels. To preserve more detailed features, the generator adopts a residual encoding–decoding network without down-sampling and up-sampling layers, which effectively competes with a residual discriminator, encouraging the accuracy of scene identification. For the training strategy, a transfer-learning-based method was applied. Instead of using either simulated or scarce real data to train the deep network, adequate simulated pairs were used to train the YUV-GAN at first. Then, pre-trained convolutional layers were optimized by real pairs to encourage the applicability of the model to real cloudy images. Qualitative and quantitative results on RICE1 and Sentinel-2A datasets confirmed that our YUV-GAN achieved state-of-the-art performance compared with other approaches. Additionally, our method combining the YUV-GAN with a transfer-learning-based training strategy led to better performance in the case of scarce training data.

Single image super-resolution (SISR) has been widely studied in recent years as a crucial technique for remote sensing applications. In this paper, a dense residual generative adversarial network (DRGAN)-based SISR method is proposed to promote the resolution of remote sensing images. Different from previous super-resolution (SR) approaches based on generative adversarial networks (GANs), the novelty of our method mainly lies in the following factors. First, we made a breakthrough in terms of network architecture to improve performance. We designed a dense residual network as the generative network in GAN, which can make full use of the hierarchical features from low-resolution (LR) images. We also introduced a contiguous memory mechanism into the network to take advantage of the dense residual block. Second, we modified the loss function and altered the model of the discriminative network according to the Wasserstein GAN with a gradient penalty (WGAN-GP) for stable training. Extensive experiments were performed using the NWPU-RESISC45 dataset, and the results demonstrated that the proposed method outperforms state-of-the-art methods in terms of both objective evaluation and subjective perspective.

Target detection is one of the important applications in the field of remote sensing. The Gaofen-3 (GF-3) Synthetic Aperture Radar (SAR) satellite launched by China is a powerful tool for maritime monitoring. This work aims at detecting ships in GF-3 SAR images using a new land masking strategy, the appropriate model for sea clutter and a neural network as the discrimination scheme. Firstly, the fully convolutional network (FCN) is applied to separate the sea from the land. Then, by analyzing the sea clutter distribution in GF-3 SAR images, we choose the probability distribution model of Constant False Alarm Rate (CFAR) detector from K-distribution, Gamma distribution and Rayleigh distribution based on a tradeoff between the sea clutter modeling accuracy and the computational complexity. Furthermore, in order to better implement CFAR detection, we also use truncated statistic (TS) as a preprocessing scheme and iterative censoring scheme (ICS) for boosting the performance of detector. Finally, we employ a neural network to re-examine the results as the discrimination stage. Experiment results on three GF-3 SAR images verify the effectiveness and efficiency of this approach.

Tremendous progress has been made in object recognition with deep convolutional neural networks (CNNs), thanks to the availability of large-scale annotated dataset. With the ability of learning highly hierarchical image feature extractors, deep CNNs are also expected to solve the Synthetic Aperture Radar (SAR) target classification problems. However, the limited labeled SAR target data becomes a handicap to train a deep CNN. To solve this problem, we propose a transfer learning based method, making knowledge learned from sufficient unlabeled SAR scene images transferrable to labeled SAR target data. We design an assembled CNN architecture consisting of a classification pathway and a reconstruction pathway, together with a feedback bypass additionally. Instead of training a deep network with limited dataset from scratch, a large number of unlabeled SAR scene images are used to train the reconstruction pathway with stacked convolutional auto-encoders (SCAE) at first. Then, these pre-trained convolutional layers are reused to transfer knowledge to SAR target classification tasks, with feedback bypass introducing the reconstruction loss simultaneously. The experimental results demonstrate that transfer learning leads to a better performance in the case of scarce labeled training data and the additional feedback bypass with reconstruction loss helps to boost the capability of classification pathway.

This study aims to detect vessels with lengths ranging from about 70 to 300 m, in Gaofen-3 (GF-3) SAR images with ultrafine strip-map (UFS) mode as fast as possible. Based on the analysis of the characteristics of vessels in GF-3 SAR imagery, an effective vessel detection method is proposed in this paper. Firstly, the iterative constant false alarm rate (CFAR) method is employed to detect the potential ship pixels. Secondly, the mean-shift operation is applied on each potential ship pixel to identify the candidate target region. During the mean-shift process, we maintain a selection matrix recording which pixels can be taken, and these pixels are called as the valid points of the candidate target. The l 1 norm regression is used to extract the principal axis and detect the valid points. Finally, two kinds of false alarms, the bright line and the azimuth ambiguity, are removed by comparing the valid area of the candidate target with a pre-defined value and computing the displacement between the true target and the corresponding replicas respectively. Experimental results on three GF-3 SAR images with UFS mode demonstrate the effectiveness and efficiency of the proposed method.

In this paper a non-parametric model based on Wasserstein CNN is proposed for color correction. It is suitable for large-scale remote sensing image preprocessing from multiple sources under various viewing conditions, including illumination variances, atmosphere disturbances, and sensor and aspect angles. Color correction aims to alter the color palette of an input image to a standard reference which does not suffer from the mentioned disturbances. Most of current methods highly depend on the similarity between the inputs and the references, with respect to both the contents and the conditions, such as illumination and atmosphere condition. Segmentation is usually necessary to alleviate the color leakage effect on the edges. Different from the previous studies, the proposed method matches the color distribution of the input dataset with the references in a probabilistic optimal transportation framework. Multi-scale features are extracted from the intermediate layers of the lightweight CNN model and are utilized to infer the undisturbed distribution. The Wasserstein distance is utilized to calculate the cost function to measure the discrepancy between two color distributions. The advantage of the method is that no registration or segmentation processes are needed, benefiting from the local texture processing potential of the CNN models. Experimental results demonstrate that the proposed method is effective when the input and reference images are of different sources, resolutions, and under different illumination and atmosphere conditions.