Search Results (759)

Synthetic aperture radar (SAR) image classification under limited data conditions faces two major challenges: inter-class similarity, where distinct radar targets (e.g., tanks and armored trucks) have nearly identical scattering characteristics, and intra-class variability, caused by speckle noise, pose changes, and differences in depression angle. To address these challenges, we propose MHD-ProtoNet, a meta-learning framework that extends prototypical networks with two key innovations: margin-aware hard example mining to better separate confusable classes by enforcing prototype distance margins, and dual-loss optimization to refine embeddings and improve robustness to noise-induced variations. Evaluated on the MSTAR dataset in a five-way one-shot task, MHD-ProtoNet achieves $76.80\%$ accuracy, outperforming the Hybrid Inference Network (HIN) $(74.70\%)$, as well as standard few-shot methods such as prototypical networks $(69.38\%)$, ST-PN $(72.54\%)$, and graph-based models like ADMM-GCN $(61.79\%)$ and DGP-NET $(68.60\%)$. By explicitly mitigating inter-class ambiguity and intra-class noise, the proposed model enables robust SAR target recognition with minimal labeled data. Full article

Although SAR systems can provide high-resolution aerial view images all-day, all-weather, the aspect and pose-sensitivity of the SAR target signatures, which defies the Gestalt perceptual principles, sets a frustrating performance upper bound for SAR Automatic Target Recognition (ATR). Therefore, we propose a network to support context-guided ATR by using aligned Electro-Optical (EO)-SAR image pairs. To realize EO-SAR image scene grammar alignment, the stable context features highly correlated to the parking patterns of the vehicle and aircraft targets are extracted from the EO images as prior knowledge, which is used to assist SAR-ATR. The proposed network consists of a Scene Recognition Module (SRM) and an instance-level Cross-modality ATR Module (CATRM). The SRM is based on a novel light-condition-driven adaptive EO-SAR decision weighting scheme, and the Outlier Exposure (OE) approach is employed for SRM training to realize Out-of-Distribution (OOD) scene detection. Once the scene depicted in the cut of interest is identified with the SRM, the image cut is sent to the CATRM for ATR. Considering that the EO-SAR images acquired from diverse observation angles often feature unbalanced quality, a novel class-incremental learning method based on the Context-Guided Re-Identification (ReID)-based Key-view (CGRID-Key) exemplar selection strategy is devised so that the network is capable of continuous learning in the open-world deployment environment. Vehicle ATR experimental results based on the UNICORN dataset, which consists of 360-degree EO-SAR images of an army base, show that the CGRID-Key exemplar strategy offers a classification accuracy 29.3% higher than the baseline model for the incremental vehicle category, SUV. Moreover, aircraft ATR experimental results based on the aligned EO-SAR images collected over several representative airports and the Arizona aircraft boneyard show that the proposed network achieves an F1 score of 0.987, which is 9% higher than YOLOv8. Full article

Crop classification plays a vital role in acquiring the spatial distribution of agricultural crops, enhancing agricultural management efficiency, and ensuring food security. With the continuous advancement of remote sensing technologies, achieving efficient and accurate crop classification using remote sensing imagery has become a prominent research focus. Conventional approaches largely rely on empirical rules or single-feature selection (e.g., NDVI or VV) for temporal feature extraction, lacking systematic optimization of multimodal feature combinations from optical and radar data. To address this limitation, this study proposes a crop classification method based on feature-level fusion of multimodal remote sensing data, integrating the complementary advantages of optical and SAR imagery to overcome the temporal and spatial representation constraints of single-sensor observations. The study was conducted in Story County, Iowa, USA, focusing on the growth cycles of corn and soybean. Eight vegetation indices (including NDVI and NDRE) and five polarimetric features (VV and VH) were constructed and analyzed. Using a random forest algorithm to assess feature importance, NDVI+NDRE and VV+VH were identified as the optimal feature combinations. Subsequently, 16 scenes of optical imagery (Sentinel-2) and 30 scenes of radar imagery (Sentinel-1) were fused at the feature level to generate a multimodal temporal feature image with 46 channels. Using Cropland Data Layer (CDL) samples as reference data, a U-Net deep neural network was employed for refined crop classification and compared with single-modal results. Experimental results demonstrated that the fusion model outperforms single-modal approaches in classification accuracy, boundary delineation, and consistency, achieving training, validation, and test accuracies of 95.83%, 91.99%, and 90.81% respectively. Furthermore, consistent improvements were observed across evaluation metrics, including F1-score, precision, and recall. Full article

This paper proposes a multi-scale time-frequency representation fusion network (MTRFN) for target recognition in synthetic aperture radar (SAR) imagery. Leveraging the spectral characteristics of six radar sub-views, the model incorporates a multi-scale representation fusion (MRF) module to extract discriminative frequency-domain features from two types of radar sub-views with high learnability. Additionally, physical scattering characteristics in SAR images are captured via time-frequency domain analysis. To enhance feature integration, a gated fusion network performs adaptive feature concatenation. The MRF module integrates a lightweight residual block to reduce network complexity and employs a coordinate attention mechanism to prioritize salient targets in the frequency spectrum over background noise, aligning the model’s focus with physical scattering principles. Furthermore, the model introduces an angular additive margin loss function during classification to enhance intra-class compactness and inter-class separability while reducing computational overhead. Compared with existing interpretable methods, the proposed approach combines architectural transparency with physical interpretability, thereby lowering the risk of recognition errors. Extensive experiments conducted on four public datasets demonstrate that the proposed MTRFN significantly outperforms existing benchmark methods. Comparative experiments using heat maps further confirm that the proposed physical feature-guided module effectively directs the model’s attention toward the target rather than the background. Full article

Polarimetric Synthetic Aperture Radar (PolSAR) image classification is one of the most important applications in remote sensing. The impressive superpixel generation approaches can improve the efficiency of the subsequent classification task and restrain the influence of the speckle noise to an extent. Most of the classical PolSAR superpixel generation approaches use the features extracted manually and even only consider the pseudocolor images. They do not make full use of polarimetric information and do not necessarily lead to good enough superpixels. The deep learning methods can extract effective deep features but they are difficult to combine with superpixel generation to achieve true end-to-end training. Addressing the above issues, this study proposes an end-to-end fully convolutional superpixel generation network for PolSAR images. It integrates the extraction of polarization information features and the generation of PolSAR superpixels into one step. PolSAR superpixels can be generated based on deep polarization feature extraction and need no traditional clustering process. Both the performance and efficiency of generations of PolSAR superpixels can be enhanced effectively. The experimental results on various PolSAR datasets show that the proposed method can achieve impressive superpixel segmentation by fitting the real boundaries of different types of ground objects effectively and efficiently. It can achieve excellent classification performance by connecting a very simple classification network, which is helpful to improve the efficiency of the subsequent PolSAR image classification tasks. Full article

Floods stand as one of the most harmful natural disasters, which have become more dangerous because of climate change effects on urban structures and agricultural fields. This research presents a comprehensive flood mapping approach that combines multi-sensor satellite data with a machine learning method to evaluate the July 2021 flood in the Netherlands. The research developed 25 different feature scenarios through the combination of Sentinel-1, Landsat-8, and Radarsat-2 imagery data by using backscattering coefficients together with optical Normalized Difference Water Index (NDWI) and Hue, Saturation, and Value (HSV) images and Synthetic Aperture Radar (SAR)-derived Grey Level Co-occurrence Matrix (GLCM) texture features. The Random Forest (RF) classifier was optimized before its application based on two different flood-prone regions, which included Zutphen’s urban area and Heijen’s agricultural land. Results demonstrated that the multi-sensor fusion scenarios (S18, S20, and S25) achieved the highest classification performance, with overall accuracy reaching 96.4% (Kappa = 0.906–0.949) in Zutphen and 87.5% (Kappa = 0.754–0.833) in Heijen. For the flood class F1 scores of all scenarios, they varied from 0.742 to 0.969 in Zutphen and from 0.626 to 0.969 in Heijen. Eventually, the addition of SAR texture metrics enhanced flood boundary identification throughout both urban and agricultural settings. Radarsat-2 provided limited benefits to the overall results, since Sentinel-1 and Landsat-8 data proved more effective despite being freely available. This study demonstrates that using SAR and optical features together with texture information creates a powerful and expandable flood mapping system, and RF classification performs well in diverse landscape settings. Full article

Compared to traditional real-valued neural networks, which process only amplitude information, complex-valued neural networks handle both amplitude and phase information, leading to superior performance in polarimetric synthetic aperture radar (PolSAR) image classification tasks. This paper proposes a multi-scale feature extraction (MSFE) method based on a 3D complex-valued network to improve classification accuracy by fully leveraging multi-scale features, including phase information. We first designed a complex-valued three-dimensional network framework combining complex-valued 3D convolution (CV-3DConv) with complex-valued squeeze-and-excitation (CV-SE) modules. This framework is capable of simultaneously capturing spatial and polarimetric features, including both amplitude and phase information, from PolSAR images. Furthermore, to address robustness degradation from limited labeled samples, we introduced a multi-scale learning strategy that jointly models global and local features. Specifically, global features extract overall semantic information, while local features help the network capture region-specific semantics. This strategy enhances information utilization by integrating multi-scale receptive fields, complementing feature advantages. Extensive experiments on four benchmark datasets demonstrated that the proposed method outperforms various comparison methods, maintaining high classification accuracy across different sampling rates, thus validating its effectiveness and robustness. Full article

Artificial intelligence (AI) techniques—ranging from hybrid mechanistic–machine learning (ML) ensembles to gradient-boosted decision trees, support-vector machines, and deep neural networks—are transforming the management of seasonal influenza, respiratory syncytial virus (RSV), human immunodeficiency virus (HIV), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Symptom-based triage models using eXtreme Gradient Boosting (XGBoost) and Random Forests, as well as imaging classifiers built on convolutional neural networks (CNNs), have improved diagnostic accuracy across respiratory infections. Transformer-based architectures and social media surveillance pipelines have enabled real-time monitoring of COVID-19. In HIV research, support-vector machines (SVMs), logistic regression, and deep neural network (DNN) frameworks advance viral-protein classification and drug-resistance mapping, accelerating antiviral and vaccine discovery. Despite these successes, persistent challenges remain—data heterogeneity, limited model interpretability, hallucinations in large language models (LLMs), and infrastructure gaps in low-resource settings. We recommend standardized open-access data pipelines and integration of explainable-AI methodologies to ensure safe, equitable deployment of AI-driven interventions in future viral-outbreak responses. Full article

In the task of PolSAR image classification, effectively utilizing convolutional neural networks and vision transformer models with limited labeled data poses a critical challenge. This article proposes a novel method for PolSAR image classification that combines channel-wise convolutional local perception, detachable self-attention, and a residual feedforward network. Specifically, the proposed method comprises several key modules. In the channel-wise convolutional local perception module, channel-wise convolution operations enable accurate extraction of local features from different channels of PolSAR images. The local residual connections further enhance these extracted features, providing more discriminative information for subsequent processing. Additionally, the detachable self-attention mechanism plays a pivotal role: it facilitates effective interaction between local and global information, enabling the model to comprehensively perceive features across different scales, thereby improving classification accuracy and robustness. Subsequently, replacing the conventional feedforward network with a residual feedforward network that incorporates residual structures aids the model in better representing local features, further enhances the capability of cross-layer gradient propagation, and effectively alleviates the problem of vanishing gradients during the training of deep networks. In the final classification stage, two fully connected layers with dropout prevent overfitting, while softmax generates predictions. The proposed method was validated on the AIRSAR Flevoland, RADARSAT-2 San Francisco, and RADARSAT-2 Xi’an datasets. The experimental results demonstrate that the proposed method can attain a high level of classification performance even with a limited amount of labeled data, and the model is relatively stable. Furthermore, the proposed method has lower computational costs than comparative methods. Full article

The Medium Earth Orbit (MEO) spaceborne Synthetic Aperture Radar (SAR) has great coverage ability, which can improve maritime ship target surveillance performance significantly. However, due to the huge computational load required for imaging processing and the severe defocusing caused by ship motions, traditional ship recognition conducted in focused image domains cannot process MEO SAR data efficiently. To address this issue, a multi-level focusing-classification strategy for MEO SAR ship recognition is proposed, which is applied to the range-compressed ship data domain. Firstly, global fast coarse-focusing is conducted to compensate for sailing motion errors. Then, a coarse-classification network is designed to realize major target category classification, based on which local region image slices are extracted. Next, fine-focusing is performed to correct high-order motion errors, followed by applying fine-classification applied to the image slices to realize final ship classification. Equivalent MEO SAR ship images generated by real LEO SAR data are utilized to construct training and testing datasets. Simulated MEO SAR ship data are also used to evaluate the generalization of the whole method. The experimental results demonstrate that the proposed method can achieve high classification precision. Since only local region slices are used during the second-level processing step, the complex computations induced by fine-focusing for the full image can be avoided, thereby significantly improving overall efficiency. Full article

Deep learning models have obvious advantages in detecting oil spills, but the training of deep learning models heavily depends on a large number of samples of high quality. However, due to the accidental nature, unpredictability, and urgency of oil spill incidents, it is difficult to obtain a large number of labeled samples in real oil spill monitoring scenarios. Surprisingly, few-shot learning can achieve excellent classification performance with only a small number of labeled samples. In this context, a new cross-domain few-shot SAR oil spill detection network is proposed in this paper. Significantly, the network is embedded with a hybrid attention feature extraction block, which consists of a coordinate attention module to perceive the channel information and spatial location information, as well as a global self-attention transformer module capturing the global dependencies and a multi-scale self-attention module depicting the local detailed features, thereby achieving deep mining and accurate characterization of image features. In addition, to address the problem that it is difficult to distinguish between the suspected oil film in seawater and real oil film using few-shot due to the small difference in features, this paper proposes a double loss function category determination block, which consists of two parts: a well-designed category-perception loss function and a traditional cross-entropy loss function. The category-perception loss function optimizes the spatial distribution of sample features by shortening the distance between similar samples while expanding the distance between different samples. By combining the category-perception loss function with the cross-entropy loss function, the network’s performance in discriminating between real and suspected oil films is thus maximized. The experimental results effectively demonstrate that this study provides an effective solution for high-precision oil spill detection under few-shot conditions, which is conducive to the rapid identification of oil spill accidents. Full article

The COVID-19 pandemic has presented significant challenges to global healthcare, bringing out the urgent need for reliable diagnostic tools. Computed Tomography (CT) scans have proven instrumental in detecting COVID-19-induced lung abnormalities. This study introduces Convolutional Neural Network, Graph Neural Network, and Vision Transformer (ViTGNN), an advanced hybrid model designed to enhance SARS-CoV-2 detection by combining Graph Neural Networks (GNNs) for feature extraction with Vision Transformers (ViTs) for classification. Using the strength of CNN and GNN to capture complex relational structures and the ViT capacity to classify global contexts, ViTGNN achieves a comprehensive representation of CT scan data. The model was evaluated on a SARS-CoV-2 CT scan dataset, demonstrating superior performance across all metrics compared to baseline models. The model achieved an accuracy of 95.98%, precision of 96.07%, recall of 96.01%, F1-score of 95.98%, and AUC of 98.69%, outperforming existing approaches. These results indicate that ViTGNN is an effective diagnostic tool that can be applied beyond COVID-19 detection to other medical imaging tasks. Full article

A timely and accurate distribution of crop types and cropping patterns provides a crucial reference for the management of agriculture and food security. However, accurately mapping crop types and cropping patterns in subtropical hilly and mountainous areas often face challenges such as mixed pixels resulted from fragmented patches and difficulty in obtaining optical satellites due to a frequently cloudy and rainy climate. Here we propose a crop type and cropping pattern mapping framework in subtropical hilly and mountainous areas, considering multiple sources of satellites (i.e., Landsat 8/9, Sentinel-2, and Sentinel-1 images and GF 1/2/7). To develop this framework, six types of variables from multi-sources data were applied in a random forest classifier to map major summer crop types (singe-cropped rice and double-cropped rice) and winter crop types (rapeseed). Multi-scale segmentation methods were applied to improve the boundaries of the classified results. The results show the following: (1) Each type of satellite data has at least one variable selected as an important feature for both winter and summer crop type classification. Apart from the endmember variables, the other five extracted variable types are selected by the RF classifier for both winter and summer crop classifications. (2) SAR data can capture the key information of summer crops when optical data is limited, and the addition of SAR data can significantly improve the accuracy as to summer crop types. (3) The overall accuracy (OA) of both summer and winter crop type mapping exceeded 95%, with clear and relatively accurate cropland boundaries. Area evaluation showed a small bias in terms of the classified area of rapeseed, single-cropped rice, and double-cropped rice from statistical records. (4) Further visual examination of the spatial distribution showed a better performance of the classified crop types compared to three existing products. The results suggest that the proposed method has great potential in accurately mapping crop types in a complex subtropical planting environment. Full article

Current synthetic aperture radar (SAR) ship datasets exhibit a notable disparity in the distribution of large, medium, and small ship targets. This imbalance makes it difficult for a relatively small number of large and medium-sized ships to be effectively trained, resulting in many false alarms. Therefore, to address the issues of scale diversity, intra-class imbalance in ship data, and the feature conflict problem associated with traditional coupled detection heads, we propose an SAR image multiscale task-decoupled oriented ship target detector based on a size-aware balanced strategy. First, the multiscale target features are extracted using the multikernel heterogeneous perception module (MKHP). Meanwhile, the triple-attention module is introduced to establish the remote channel dependence to alleviate the issue of small target feature annihilation, which can effectively enhance the feature characterization ability of the model. Second, given the differences in the demand for feature information between the detection and classification tasks, a channel attention-based task decoupling dual-head (CAT2D) detector head structure is introduced to address the inherent conflict between classification and localization tasks. Finally, a new size-aware balanced (SAB) loss strategy is proposed to guide the network in focusing on the scarce targets in training to alleviate the intra-class imbalance problem during the training process. The ablation experiments on SSDD+ reflect the contribution of each component, and the results of the comparison experiments on the RSDD-SAR and HRSID datasets show that the proposed method achieves state-of-the-art performance compared to other state-of-the-art detection models. Furthermore, our approach exhibits superior detection coverage for both offshore and inshore scenarios for ship detection tasks. Full article

Soil salinization poses a significant threat to agricultural productivity, food security, and ecological sustainability in arid and semi-arid regions. Effectively and timely mapping of different degrees of salinized soils is essential for sustainable land management and ecological restoration. Although deep learning (DL) methods have been widely employed for soil salinization extraction from remote sensing (RS) data, the integration of multi-source RS data with DL methods remains challenging due to issues such as limited data availability, speckle noise, geometric distortions, and suboptimal data fusion strategies. This study focuses on the Keriya Oasis, Xinjiang, China, utilizing RS data, including Sentinel-2 multispectral and GF-3 full-polarimetric SAR (PolSAR) images, to conduct soil salinization classification. We propose a Dual-Modal deep learning network for Soil Salinization named DMSSNet, which aims to improve the mapping accuracy of salinization soils by effectively fusing spectral and polarimetric features. DMSSNet incorporates self-attention mechanisms and a Convolutional Block Attention Module (CBAM) within a hierarchical fusion framework, enabling the model to capture both intra-modal and cross-modal dependencies and to improve spatial feature representation. Polarimetric decomposition features and spectral indices are jointly exploited to characterize diverse land surface conditions. Comprehensive field surveys and expert interpretation were employed to construct a high-quality training and validation dataset. Experimental results indicate that DMSSNet achieves an overall accuracy of 92.94%, a Kappa coefficient of 79.12%, and a macro F1-score of 86.52%, positively outperforming conventional DL models (ResUNet, SegNet, DeepLabv3+). The results confirm the superiority of attention-guided dual-branch fusion networks for distinguishing varying degrees of soil salinization across heterogeneous landscapes and highlight the value of integrating Sentinel-2 optical and GF-3 PolSAR data for complex land surface classification tasks. Full article