Search Results (500)

Soil salinization in arid oases constrains soil functioning and crop production, making spatially explicit monitoring important for land management. Multispectral optical remote sensing enables large-area salinity assessment, but in oasis environments such as the Keriya Oasis, its performance can be limited by spectral confusion between salt crusts and bright bare soils, sparse vegetation cover, and strong surface heterogeneity. Synthetic aperture radar (SAR), by contrast, provides all-weather imaging capability and sensitivity to surface scattering and dielectric-related conditions, but its salinity interpretation is often affected by surface complexity and environmental coupling. To address these, a spectral index–polarimetric scattering integration framework that combines RADARSAT-2 and Landsat-8 OLI features within a simple two-dimensional (2D) feature space was developed. Two groups of models were constructed from variables selected through a data-driven screening process: (1) polarimetric feature space models based on combinations such as VanZyl volume scattering with Pauli odd-bounce or Touzi alpha scattering; and (2) multi-source feature space models that integrate the optimal polarimetric component with key spectral indicators such as SI4 and MSAVI. Among all tested models, VanZyl_vol-SI4 achieved the best performance (fitting: R² = 0.749, RMSE = 5.798 dS m⁻¹, MAE = 4.086 dS m⁻¹; validation: R² = 0.716, RMSE = 5.566 dS m⁻¹, MAE = 4.528 dS m⁻¹). The results indicate that integrating PolSAR scattering information with optical indices can improve salinity mapping relative to single-source feature spaces in the Keriya Oasis. The proposed 2D framework provides a concise way to compare different feature combinations and supports regional identification of salt-affected soils. Full article

In response to the problems of insufficient fusion of amplitude and phase heterogeneity features, deficient direction sensitivity modeling, and a single fusion level in the polarimetric synthetic aperture radar classification task, this paper proposes a PolSAR classification method based on dynamic weight adjustment and heterogeneous feature fusion. This method utilizes a dual-branch parallel structure to extract polarization features and landcover amplitude-phase direction difference features separately and constructs a three-level progressive fusion strategy of sub-branch, cross-branch, and decision layer to achieve adaptive complementation of heterogeneous features. Experiments on three standard datasets show that the classification accuracy and visual consistency of this method are significantly superior to the classical methods, with the overall accuracy being improved by 1.5% to 2.4%. Full article

The aim of the current study was to select and test the appropriate model and input parameters for remote sensing retrieval of surface soil moisture (SSM) in the case of bare Chernozems on flat and sloping terrains in northern Bulgaria under different tillage systems. Normalized synthetic aperture radar (SAR) measurements from Sentinel-1 C-band dual-pol products (Gamma-Nought in VV, ratio) were utilized in two ways to delineate SSM from environmental factors that bias determination. The accuracy of the obtained SSM prediction was evaluated against ground-based volumetric water content (VWC) measured in the 0–3.8 cm soil layer at multiple points using a TDR meter. The TDR VWC data were preliminarily calibrated against gravimetric measurements in the 0–5 cm soil layer. The obtained data for soil water retention curves in all studied variants were used to determine the range of soil moisture variation. The measured ground-based data for surface roughness generally correlate with the co-pol Gamma-Nought in VV. The data modeled with the surface soil moisture script in Sentinel Hub (SSM-SH) was calibrated using the ground-based data. Incidence angle normalization of Sentinel-1 products improved the relationship between SAR observables and SSM, when expressed as the ratio of soil moisture to total porosity (rVWC). The modeling indicated the highest importance of the optical indices, together with the temporal differences of radar descriptors sensitive to variations in soil moisture over time. Although the applied Random Forest Regression (RFR) model achieved higher accuracy during training (nRMSE of 7.27%, R² of 0.86), the Gaussian Process Regression (GPR) model provided better generalization performance on the independent validation dataset. The results proved the advantages of the joint utilization of temporal Sentinel-1 SAR measurements with Sentinel-2 optical acquisitions to determine SSM in different bare soil conditions for achieving high accuracy. Full article

Model-based polarimetric SAR (PolSAR) algorithms for bio- and geophysical parameter estimation rely on the effective separation of the combined scattering response of vegetation canopies and the soil surface through physically based models. However, the interpretation of polarimetric features derived from physical models is still subject to some ambiguity. Another strategy for complementing the model-based approaches for scattering mechanisms characterisation deals with the separation of the polarised and depolarised contributions of the PolSAR data according to their degree of polarisation. In this paper, we propose a two-component decomposition for estimating the depolarised and polarised components within the target and their corresponding coherency matrices. The method requires the previous calculation of the backscattering powers given by the model-free three-component (MF3C) decomposition, which in turn relies on the 3-D Barakat degree of polarisation. This quantitative information allows us to construct an inversion algorithm to retrieve the proportion of the polarised and depolarised contributions for all the elements of the observed coherency matrix under the reflection symmetry assumption. In essence, the proposed decomposition can be regarded as an extension of the MF3C method and, as a consequence, it enables the exploitation of both model-free and model-based approaches by using a physical rationale driven by the capability of the 3-D Barakat degree of polarisation. Therefore, practical applications can benefit from this approach as the retrieval of target parameters could presumably be done in a more accurate way by directly applying existing scattering models to both components. Indoor multi-frequency datasets acquired over three vegetation samples from the European Microwave Signature Laboratory (EMSL) and P-, L-, and C-band AIRSAR images over a boreal forest in Germany have been employed for testing the proposed decomposition. Performance analysis was performed using different polarimetric tools applied to the outcomes of the two-component decomposition, namely, the eigendecomposition and the copolar cross-correlation analysis of polarised and depolarised components, as well as histograms and a correlation analysis among backscattering powers. Overall, it has been observed that the method outputs are consistent with the theoretical expectations for the depolarised and polarised scattering components for a wide range of scenarios and sensor frequencies. Full article

Forest height is a key structural parameter for characterizing forest architecture and estimating carbon storage. However, under complex terrain and heterogeneous forest conditions, Polarimetric synthetic aperture radar (PolSAR)-based forest height inversion using multi-category features still faces several challenges, including feature redundancy, insufficient characterization of the nonlinear couplings among high-dimensional features by deep learning models, and the difficulty of jointly achieving model stability and interpretability. In this paper, to address these issues, we propose a method for SHapley Additive exPlanations (SHAP) interpretability-driven PolSAR forest height inversion based on deep learning and multi-category feature fusion. Firstly, a deep neural network (DNN) is constructed, and SHAP is introduced to interpret the model decision process, enabling the identification of key feature interactions with clear physical significance and guiding the iterative model optimization in an explainability-driven manner. Furthermore, a SHAP-guided feature attention DNN is developed, in which the feature contribution scores are incorporated as prior knowledge for attention weight initialization, thereby establishing a closed-loop modeling framework from “interpretation” to “optimization”. Experiments were conducted at the site of the Huangfengqiao forest farm, Youxian County, Hunan province, China, using ALOS-2 L-band fully polarimetric SAR imagery. The experimental results demonstrated that the proposed method can significantly outperform the conventional machine learning approaches and various deep learning architectures for forest height inversion. The final model achieved a coefficient of determination (R²) score of 0.75 and a root-mean-square error (RMSE) of 1.35 m on the test dataset. These findings indicate that the combination of SHAP-driven multi-category feature fusion and deep learning can effectively enhance both the inversion accuracy and physical interpretability, providing a reliable solution for PolSAR-based forest structural parameter retrieval at the Huangfengqiao study site, with potential applicability to complex terrain conditions. Full article

To address the limitations of traditional forest above-ground biomass (AGB) retrieval methods—namely, the restricted accuracy of physical models and the limited generalization ability of purely data-driven models—this study proposes a differentiable physical modeling (DPM) approach for forest AGB estimation. The method adopts the water cloud model (WCM) as a physics-based framework, grounded in radiative transfer theory, and integrates C-band synthetic aperture radar (SAR) data with multispectral imagery. Within the PyTorch tensor computation framework, automatic differentiation (AD) is employed to seamlessly couple the WCM with the deep fully connected neural network (DFCNN), enabling a differentiable implementation of the WCM. Using mean squared error (MSE) as the loss function, the neural network parameters are optimized through backpropagation and gradient descent, thereby constructing an end-to-end trainable DPM model that effectively retrieves forest AGB while preserving physical interpretability and generalization capability. To validate the proposed method, two representative test sites were selected: Simao in Pu’er, Yunnan Province, and Genhe in Inner Mongolia. GF-3 PolSAR and RADARSAT-2 data were used to extract backscattering coefficients and compute the radar vegetation index (RVI), while Landsat 8 OLI imagery was employed to calculate the normalized difference vegetation index (NDVI), difference vegetation index (DVI), and soil-adjusted vegetation index (SAVI). These datasets, together with ASTER GDEM, field-measured biomass, and other relevant datasets, were integrated to construct a multisource dataset combining remote sensing and ground observations. The performance of the DPM model was then compared with the traditional WCM and several data-driven models, including the fully connected neural network (FNN), generalized regression neural network (GRNN), RF, and Adaptive Boosting (AdaBoost). The results indicate that the DPM model achieved R² = 0.60, RMSE = 24.23 Mg/ha, Bias = 0.4 Mg/ha, and ubRMSE = 22.43 Mg/ha in Simao, and R² = 0.48, RMSE = 33.29 Mg/ha, Bias = 0.87 Mg/ha, and ubRMSE = 33.28 Mg/ha in Genhe, demonstrating consistently better performance than both the WCM and all tested data-driven models. The DPM model demonstrated consistent performance across ecologically contrasting forest regions. It alleviated the systematic overestimation bias of purely data-driven models and overcame the limitations in predictive accuracy resulting from the simplified structure of the WCM. The differentiability of the WCM enables the loss function errors to be backpropagated through the neural network, thereby allowing the optimization of the physical model parameters. Overall, the DPM framework integrates the advantages of both physical models and data-driven approaches, providing an estimation method with acceptable accuracy for forest AGB retrieval. It also offers theoretical and practical insights for the integration of deep learning and physical knowledge in other research fields. Full article

Forest height is a key parameter for quantifying forest biomass and carbon stocks and serves as an important indicator of forest ecosystem health. The successful launch of the European Space Agency’s P-band Biomass satellite, which provides Polarimetric Interferometric Synthetic Aperture Radar (PolInSAR) data for global high-precision forest height mapping, heralds a new era in global forest carbon monitoring. However, the accuracy of forest height inversion is significantly influenced by scattering mechanisms. This study investigates the impact of dominant scattering mechanisms on forest height inversion accuracy. Four classical algorithms were selected: the polarimetric phase center height estimation method (PPC), the complex coherence phase center differencing algorithm (CCPCD), the coherence amplitude inversion method (CAI), and the hybrid inversion method using both phase and coherence information. The Freeman–Durden three-component decomposition was employed to identify the dominant scattering mechanisms. The results show that (1) at P-band, inversion model performance exhibits strong coupling with scattering mechanisms, and no single algorithm achieves global robustness; (2) the hybrid inversion method using both phase and coherence information performs better in regions dominated by surface and double-bounce scattering, whereas the coherence amplitude inversion method (CAI) yields higher accuracy in volume-scattering-dominated regions; and (3) the adaptive joint inversion strategy based on scattering mechanisms achieved a root mean square error ($\mathrm{RMSE}$) of 4.62 m and a coefficient of determination ($R^2$) of 0.76 at P-band, representing an improvement of approximately 30% over the best single-model performance ($\mathrm{RMSE}$ = 6.51 m). This approach overcomes the accuracy limitations of single models in complex global forest scenarios and provides a valuable reference for scientific forest height inversion. Full article

Accurate large-scale sea ice classification is vital for Arctic maritime activities. However, this task faces a fundamental challenge. Operational surveillance is restricted to wide-swath dual-polarization data, which limits classification precision due to polarimetric information deficiency. Conversely, while the quad-polarization mode offers the comprehensive scattering details required for more accurate classification, its narrow swath width prevents efficient large-scale monitoring. To address this challenge, we propose an enhanced sea ice classification method relying on limited co-region quad-polarization information to enhance dual-polarization data classification accuracy across larger spatiotemporal scales. Specifically, we construct a polarization-guided cross-mode knowledge distillation framework featuring an asymmetric teacher–student architecture. In this design, a hybrid CNN-Transformer teacher extracts robust scattering features from quad-polarization data to guide a lightweight student network operating on dual-polarization inputs. Through this transfer, the student acquires rich feature representations comparable to quad-polarization data, effectively compensating for the missing polarimetric scattering information. Experimental results on GF3-02 data demonstrate that the proposed method significantly outperforms the standalone dual-polarization network baseline, achieving an overall accuracy of 86.15%. This validates the effectiveness of the proposed method in enabling high-precision sea ice classification for large-scale monitoring. Full article

Target decomposition is an essential method for the interpretation of polarimetric Synthetic Aperture Radar (SAR). Most current polarimetric target decomposition methods are designed for quad-pol SAR data, while there is a scarcity of methods tailored for dual-pol SAR data, and these methods often struggle to accurately capture the complete scattering components of targets. Compared to quad-pol SAR, space-borne SAR systems more frequently acquire dual-pol SAR data, which offers a wider observation swath and higher resolution. The fast generalized polarimetric target decomposition (FGPTD) method has exhibited excellent target decomposition performance for quad-pol SAR data by searching for the optimal scattering models through nonlinear optimization. To address the core problem of inaccurate scattering component extraction in dual-pol SAR, deep learning is adopted to simulate the nonlinear optimization process of the FGPTD method. Its powerful nonlinear mapping capability enables the model to learn the intrinsic correlation between dual-pol SAR data and the complete scattering components obtained by FGPTD. Therefore, this paper proposes a model and learning-aided target decomposition method for dual-pol SAR. Firstly, FGPTD is performed on existing quad-pol SAR data. Subsequently, a mapping set between dual-pol SAR data and scattering components is constructed. Then, a neural network that integrates residual connections and dilated convolutional kernels is trained using the constructed mapping set. Finally, the well-trained neural network is tested on dual-pol SAR data from other regions and other sensors. Experimental results demonstrate that the proposed method’s target decomposition results are close to those of quad-pol target decomposition and superior to current state-of-the-art dual-pol target decomposition methods. Full article

In many land cover classification tasks, the limited precision of individual sensors hinders the accurate separation of certain classes, largely due to the complexity of the Earth’s surface morphology. To mitigate these issues, decision fusion methodologies are employed, allowing data from multiple sensors to be synthesized into robust and more conclusive classification outcomes. This study employs fully polarimetric Synthetic Aperture Radar (PolSAR) imagery and leverages the strengths of three decomposition methods, namely Pauli’s, Krogager’s, and Cloude’s, by extracting their respective components for improved detection. From each decomposition method, three scattering components are derived, enabling the extraction of informative features that describe the scattering behavior associated with various land cover types. The extracted scattering features, treated as independent sensors, were used to train three neural network classifiers. The resulting outputs were then considered as local decisions for each land cover type and subsequently fused through a decision fusion rule to generate more complete and accurate classification results. Experimental results demonstrate that the proposed Multi-Class Bahadur–Lazarsfeld Expansion (MC-BLE) fusion significantly enhances classification performance, achieving an overall accuracy (OA) of 95.78% and a Kappa coefficient of 0.94. Compared to individual classification methods, the fusion notably improved per-class accuracy, particularly for complex land cover boundaries. The core innovation of this work is the transformation of the Bahadur–Lazarsfeld Expansion (BLE), originally designed for binary decision fusion into a multi-class framework capable of addressing multiple land cover types, resulting in a more effective and reliable decision fusion strategy. Full article

Complex-valued convolutional neural networks (CVCNNs) have demonstrated strong capabilities for polarimetric synthetic aperture radar (PolSAR) image classification by effectively integrating both amplitude and phase information inherent in polarimetric data. However, their practical deployment faces significant challenges due to high computational costs and performance degradation caused by extremely limited labeled samples. To address these challenges, a lightweight CV Siamese network (LCVSNet) is proposed for few-shot PolSAR image classification. Considering the constraints of limited hardware resources in practical applications, simple one-dimensional (1D) CV convolutions along the scattering dimension are combined with two-dimensional (2D) lightweight CV convolutions. In this way, the inter-element dependencies of polarimetric coherency matrix and the spatial correlations between neighboring units can be captured effectively, while simultaneously reducing computational costs. Furthermore, LCVSNet incorporates a contrastive learning (CL) projection head to explicitly optimize the feature space. This optimization can effectively enhance the feature discriminability, leading to accurate classification with a limited number of labeled samples. Experiments on three real PolSAR datasets demonstrate the effectiveness and practical utility of LCVSNet for PolSAR image classification with a small number of labeled samples. Full article

Deep learning has significantly advanced PolSAR image processing, with a growing trend of integrating mathematical theories into deep neural networks to enhance their capabilities with regard to complex data. Kolmogorov–Arnold networks (KANs), which leverage nonlinear mappings derived from the Kolmogorov–Arnold theorem for automatic feature extraction, present a promising yet underexplored direction for PolSAR image classification. However, existing real-valued KAN-based layers fall short in effectively exploiting the complex-valued characteristics of PolSAR data, overlooking the important phase information. In this paper, we propose a complex-valued convolutional Kolmogorov–Arnold framework for PolSAR image classification (CV-CPKAN). The framework introduces complex KAN convolution layers, which are further employed to construct a multi-branch complex KAN convolution (MBComplexKConv) block, effectively extracting multi-scale features from both the amplitude and phase components of PolSAR data. Additionally, a complex-valued variant of PolyLoss (CV-PolyLoss) is proposed as our classification loss function. Through extensive evaluations on three benchmark PolSAR datasets, CV-CPKAN consistently surpasses state-of-the-art models based on CNN, Transformer and Mamba, achieving overall accuracies of 99.86%, 99.80% and 99.74% on Flevoland, San Francisco and Oberpfaffenhofen datasets, respectively. These results underscore the effectiveness of integrating convolutions with KAN-based nonlinear mapping, providing a new avenue for further research in PolSAR image classification. Full article

Sub-meter-resolution Polarimetric Synthetic Aperture Radar (PolSAR) imagery enables precise building footprint extraction but introduces complex scattering correlated with fine spatial structures. This change renders both traditional methods, which rely on simplified scattering models, and existing deep learning approaches, which sacrifice spatial detail through multi-looking, inadequate for high-precision extraction tasks. To address this, we propose an Orthogonal Dual-Resolution Network (ODRNet) for end-to-end, precise segmentation directly from single-look complex (SLC) data. Unlike complex-valued neural networks that suffer from high computational cost and optimization difficulties, our approach decomposes complex-valued data into its orthogonal real and imaginary components, which are then concurrently fed into a Dual-Resolution Branch (DRB) with Bilateral Information Fusion (BIF) to effectively balance the trade-off between semantic and spatial details. Crucially, we introduce an auxiliary Polarization Orientation Angle (POA) regression task to enforce physical consistency between the orthogonal branches. To tackle the challenge of diverse building scales, we designed a Multi-scale Aggregation Pyramid Pooling Module (MAPPM) to enhance contextual awareness and a Pixel-attention Fusion (PAF) module to adaptively fuse dual-branch features. Furthermore, we have constructed a VHR PolSAR building footprint segmentation dataset to support related research. Experimental results demonstrate that ODRNet achieves 64.3% IoU and 78.27% F1-score on our dataset, and 73.61% IoU with 84.8% F1-score on a large-scale SLC scene, confirming the method’s significant potential and effectiveness in high-precision building extraction directly from SLC. Full article

Qarhan Salt Lake, located in the Qaidam Basin of northwestern China, is a highland lake characterized by diverse surface features, including salt lakes, salt crusts, and saline-alkali lands. Investigating the distribution and dynamic variations of salt crusts is essential for mineral resource development and regional ecological monitoring. To this end, the surface of the study area was categorized into several types according to micro-geomorphological characteristics. Polarimetric synthetic aperture radar (PolSAR), which provides rich scattering information, is well suited for distinguishing these surface categories. To achieve more accurate classification of salt crust types, the scattering differences among various types were comparatively analyzed. Stable samples were further selected using unsupervised Wishart clustering with reference to field survey results. Besides, to address the weak inter-class separability among different salt crust types, this paper proposes a PolSAR classification method tailored for salt crust discrimination by integrating unsupervised feature learning, attention-based feature optimization, and global context modeling. In this method, convolutional autoencoder (CAE) is first employed to learn discriminative local scattering representations from original polarimetric features, enabling effective characterization of subtle scattering differences among salt crust types. Vision Transformer (ViT) is introduced to model global scattering relationships and spatial context at the image-patch level, thereby improving the overall consistency of classification results. Meanwhile, the attention mechanism is used to bridge local scattering representations and global contextual information, enabling joint optimization of key scattering features. Experiments on fully polarimetric Gaofen-3 and dual-polarimetric Sentinel-1 data show that the proposed method outperforms the best competing method by $2.34\%$ and $1.17\%$ in classification accuracy, respectively. In addition, using multi-temporal Sentinel-1 data, recent temporal changes in salt crust distribution are identified and analyzed. Full article

The caldén (Neltuma caldenia) forest, a xerophytic low-stature ecosystem in central Argentina, faces increasing threats from land use change and desertification. This study assesses the capability of full-polarimetric L-band SAR data from the Argentine SAOCOM-1A satellite to characterise forest attributes in this ecosystem. We computed the Generalised Radar Vegetation Index (GRVI) and compared it with aboveground biomass and tree canopy cover data from the Second National Forest Inventory, under fire and non-fire conditions. We also assessed other SAR indices and polarimetric decompositions. GRVI values exhibited limited variability relative to the broad range of field-estimated biomass, and most regression models were not statistically significant. Nevertheless, GRVI effectively distinguished woody from non-woody vegetation and showed a weak correlation with canopy cover. Statistically significant, albeit weak, correlations were also observed between biomass and specific polarimetric components, such as the helix term of the Yamaguchi decomposition and the Pauli volume component. Key challenges included limited spatial and temporal coverage of SAOCOM-1A data and the distribution of field plots. Despite these limitations, our results support the use of GRVI for land cover monitoring in semiarid regions, emphasising the importance of multitemporal data, integration with C-band SAR, and enhanced field sampling to improve forest attribute modelling. Full article