Search Results (43)

Hyperspectral image (HSI) and light detection and ranging (LiDAR) data offer complementary spectral and structural information; however, the integration of these high-dimensional, heterogeneous modalities poses significant challenges. We propose a Global–Local Mamba dual-modality fusion framework (GL-Mamba) for HSI–LiDAR classification. Each sensor’s input is decomposed into low- and high-frequency sub-bands: lightweight 3D/2D CNNs process low-frequency spectral–spatial structures, while compact transformers handle high-frequency details. The outputs are aggregated using a global–local Mamba block, a state-space sequence model that retains local context while capturing long-range dependencies with linear complexity. A cross-attention module aligns spectral and elevation features, yielding a lightweight, efficient architecture that preserves fine textures and coarse structures. Experiments on Trento, Augsburg, and Houston2013 datasets show that GL-Mamba outperforms eight leading baselines in accuracy and kappa coefficient, while maintaining high inference speed due to its dual-frequency design. These results highlight the practicality and accuracy of our model for multimodal remote-sensing applications. Full article

The wandering reach of the Yellow River has long been a pivotal area of research due to its drastic fluctuations in water-sediment dynamics, frequent shifts in the main channel, and complex river regime evolution. Studies on the main-channel morphological evolution in this reach have focused on the analysis of parameters related to the overall oscillation or have only analyzed a certain reach within the wandering reach, with a lack of detailed studies based on the different characteristics of each area. Therefore, taking the Xiaolangdi Reservoir–Gaocun reach as the research area, by constructing a two-dimensional water-sediment dynamic model, the erosion–deposition characteristics of different sub-reaches and the morphological evolution characteristics of key cross-sections were quantified and analyzed. Based on measured hydrological, sediment, and topographic data, the temporal and spatial changes in the bankfull area and fluvial facies coefficient of typical sections before and after the construction of Xiaolangdi Reservoir were analyzed. By interpreting remote sensing images, the spatio-temporal variation characteristics of the migration distance and bending coefficient of different reaches before and after the construction of Xiaolangdi Reservoir were calculated, and the key factors influencing the evolution of river morphology parameters were identified. The results showed that after the Xiaolangdi Reservoir operation, the overall erosion of the Huayuankou–Jiahetan reach is greater than the deposition, and the erosion is more obvious in dry years. The river course direction and control engineering play a significant role in controlling the morphological evolution of the main channel during the process, causing the R2 reach to significantly swing to the north bank and the R3 reach to the south bank. When the sediment transport coefficient values were between 0 and 0.005 kg.s.m⁻⁶, water-sediment had a positive effect on shaping and evolving the main-channel morphology. The long-term low-sand discharge of Xiaolangdi Reservoir and the continuous improvement of river regulation projects are the main reasons for the above changes. The results can provide support for controlling the evolution of the main channel and improving river regulation projects. Full article

The performance of remote sensing object detectors often degrades severely when deployed in new operational environments due to covariate shift in the data distribution. Existing evaluation paradigms, which primarily rely on aggregate performance metrics such as mAP, generally lack the analytical depth to provide insights into the mechanisms behind such generalization failures. To fill this critical gap, we propose the GRADE (Generalization Robustness Assessment via Distributional Evaluation) framework, a multi-dimensional, systematic methodology for assessing model robustness. The framework quantifies shifts in background context and object-centric features through a hierarchical analysis of distributional divergence, utilizing Scene-level Fréchet Inception Distance (FID) and Instance-level FID, respectively. These divergence measures are systematically integrated with a standardized performance decay metric to form a unified, adaptively weighted Generalization Score (GS). This composite score serves not only as an evaluation tool but also as a powerful analytical tool, enabling the fine-grained attribution of performance loss to specific sources of domain shift—whether originating from scene variations or anomalies in object appearance. Compared to conventional single-dimensional evaluation methods, the GRADE framework offers enhanced interpretability, a standardized evaluation protocol, and reliable cross-model comparability, establishing a principled theoretical foundation for cross-domain generalization assessment. Extensive empirical validation on six mainstream remote sensing benchmark datasets and multiple state-of-the-art detection models demonstrates that the model rankings produced by the GRADE framework exhibit high fidelity to real-world performance, thereby effectively quantifying and explaining the cross-domain generalization penalty. Full article

Remote sensing video (RSV) provides continuous, high spatiotemporal earth observations that are increasingly important for environmental monitoring, disaster response, infrastructure inspection and urban management. Despite this potential, operational use of video streams is hindered by very large data volumes, heterogeneous acquisition platforms, inconsistent preprocessing practices, and the absence of standardized formats that deliver data ready for immediate analysis. These shortcomings force repeated low-level computation, complicate semantic extraction, and limit reproducibility and cross-sensor integration. This manuscript presents a principled multi-level analysis-ready data (ARD) model for remote sensing video, named VideoARD, along with VideoCube, a spatiotemporal management and query infrastructure that implements and operationalizes the model. VideoARD formalizes semantic abstraction at scene, object, and event levels and defines minimum and optimal readiness configurations for each level. The proposed pipeline applies stabilization, georeferencing, key frame selection, object detection, trajectory tracking, event inference, and entity materialization. VideoCube places the resulting entities into a five-dimensional structure indexed by spatial, temporal, product, quality, and semantic dimension, and supports earth observation OLAP-style operations to enable efficient slicing, aggregation, and drill down. Benchmark experiments and three application studies, covering vessel speed monitoring, wildfire detection, and near-real-time three-dimensional reconstruction, quantify system performance and operational utility. Results show that the proposed approach achieves multi-gigabyte-per-second ingestion under parallel feeds, sub-second scene retrieval for typical queries, and second-scale trajectory reconstruction for short tracks. Case studies demonstrate faster alert generation, improved detection consistency, and substantial reductions in preprocessing and manual selection work compared with on-demand baselines. The principal trade-off is an upfront cost for materialization and storage that becomes economical when queries are repeated or entities are reused. The contribution of this work lies in extending the analysis-ready data concept from static imagery to continuous video streams and in delivering a practical, scalable architecture that links semantic abstraction to high-performance spatiotemporal management, thereby improving responsiveness, reproducibility, and cross-sensor analysis for Earth observation. Full article

Chlorophyll-a (Chl-a) and chemical oxygen demand (COD) are key indicators for water quality evaluation. In previous research on the inversion of Chl-a and COD concentrations using hyperspectral data, disparities in hyperspectral data types have constrained the universality of the inversion models. To solve this problem, in this study, synchronous in situ hyperspectral data and water samples were collected from 308 stations within the river networks of Zhongshan City. Four inversion models, support vector regression (SVR), random forest (RF), backpropagation neural network (BPNN), and one-dimensional convolutional neural network (1D-CNN), were established using the original reflectance (R), remote sensing reflectance (Rrs), and their normalized forms as inputs. To evaluate the robustness of the models, their performance was assessed via cross-reflectance type validation. For example, a model was trained using R data and then tested with Rrs data. The results show that using the normalized hyperspectral data for modeling not only improves the accuracy of the inversion results of Chl-a and COD concentrations, but also effectively unifies different types of hyperspectral data, thereby improving the versatility of the inversion model. This study provides a reference for constructing a general water quality inversion model based on hyperspectral data. Full article

The satellite Internet of Things (SatIoT) enables real-time acquisition and large-scale coverage of hyperspectral imagery, providing essential data support for decision-making in domains such as geological exploration, environmental monitoring, and urban management. Hyperspectral remote sensing classification constitutes a critical component of intelligent applications driven by the SatIoT, yet it faces two major challenges: the massive data volume imposes heavy storage and processing burdens on conventional satellite systems, while dimensionality reduction often compromises classification accuracy; furthermore, mainstream neural network models are constrained by insufficient labeled data and spectral shifts, frequently leading to misclassification of unknown categories and degradation of cross-regional performance. To address these issues, this study proposes an open-set hyperspectral classification method with dual branches of reconstruction and prototype-based classification. Specifically, we build upon an autoencoder. We design a spectral–spatial attention module and an information residual connection module. These modules accurately capture spectral–spatial features. This improves the reconstruction accuracy of known classes. It also adapts to the high-dimensional characteristics of satellite data. Prototype representations of unknown classes are constructed by incorporating classification confidence, enabling effective separation in the feature space and targeted recognition of unknown categories in complex scenarios. By jointly leveraging prototype distance and reconstruction error, the proposed method achieves synergistic improvement in both accurate classification of known classes and reliable detection of unknown ones. Comparative experiments and visualization analyses on three publicly available datasets: Salinas-A, PaviaU, and Dioni-demonstrate that the proposed approach significantly outperforms baseline methods such as MDL4OW and IADMRN in terms of unknown detection rate (UDR), open-set overall accuracy (OpenOA), and open-set F1 score, while on the Salinas-A dataset, the performance gap between closed-set and open-set classification is as small as 1.82%, highlighting superior robustness. Full article

With the rapid advancement of hyperspectral remote sensing technology, the security of hyperspectral images (HSIs) has become a critical concern. However, traditional image encryption methods—designed primarily for grayscale or RGB images—fail to address the high dimensionality, large data volume, and spectral-domain characteristics inherent to HSIs. Existing chaotic encryption schemes often suffer from limited chaotic performance, narrow parameter ranges, and inadequate spectral protection, leaving HSIs vulnerable to spectral feature extraction and statistical attacks. To overcome these limitations, this paper proposes a novel hyperspectral image encryption algorithm based on a newly designed two-dimensional cross-coupled hyperchaotic map (2D-CSCM), which synergistically integrates Cubic, Sinusoidal, and Chebyshev maps. The 2D-CSCM exhibits superior hyperchaotic behavior, including a wider hyperchaotic parameter range, enhanced randomness, and higher complexity, as validated by Lyapunov exponents, sample entropy, and NIST tests. Building on this, a layered encryption framework is introduced: spectral-band scrambling to conceal spectral curves while preserving spatial structure, spatial pixel permutation to disrupt correlation, and a bit-level diffusion mechanism based on dynamic DNA encoding, specifically designed to secure high bit-depth digital number (DN) values (typically >8 bits). Experimental results on multiple HSI datasets demonstrate that the proposed algorithm achieves near-ideal information entropy (up to 15.8107 for 16-bit data), negligible adjacent-pixel correlation (below 0.01), and strong resistance to statistical, cropping, and differential attacks (NPCR ≈ 99.998%, UACI ≈ 33.30%). The algorithm not only ensures comprehensive encryption of both spectral and spatial information but also supports lossless decryption, offering a robust and practical solution for secure storage and transmission of hyperspectral remote sensing imagery. Full article

Three-dimensional (3D) ocean physical fields are essential for understanding ocean dynamics, but reconstructing them solely from sea-surface remote sensing remains challenging. We present 3DV-Unet, an end-to-end deep learning framework that reconstructs eddy-resolving three-dimensional essential ocean variables (temperature, salinity, and currents) from multi-source satellite data. The model employs a 3D Vision Transformer bottleneck to capture cross-depth and cross-variable dependencies, ensuring physically consistent reconstruction. Trained on 2011–2019 reanalysis and satellite data, 3DV-Unet achieves RMSEs of ~0.30 °C for temperature, 0.11 psu for salinity, and 0.05 m/s for currents, with all R² values above 0.93. Error analyses further indicate higher reconstruction errors in dynamically complex regions such as the Kuroshio Extension, while spectral analysis indicates good agreement at 100 km+ but systematic deviation in the 20–100 km band. Independent validation against 6113 Argo profiles confirms its ability to reproduce realistic vertical thermohaline structures. Moreover, the reconstructed 3D fields capture mesoscale eddy structures and their life cycle, offering a valuable basis for investigating ocean circulation, energy transport, and regional variability. These results demonstrate the potential of end-to-end volumetric deep learning for advancing high-resolution 3D ocean reconstruction and supporting physical oceanography and climate studies. Full article

Convolutional neural networks have made significant progress in multimodal remote sensing image classification, but traditional convolutional neural networks are limited by fixed-size convolutional kernels, which are unable to effectively model and adequately extract contextual information; hyperspectral imagery and LiDAR data have comparatively large information differences, which do not allow for effective information interaction and fusion. Based on this, this paper proposes a multimodal dual fusion network (CSTC) based on the Vision Transformer for the collaborative classification of HSI and LiDAR data. The model is designed through a two-branch architecture: the HSI branch extracts spectral–spatial features by dimensionality reduction using principal component analysis and inputs them into the cross-connectivity feature fusion module; the LiDAR branch mines spatial elevation features through the stacked MobileNetV2 module. The features of the two branches are encoded by a Transformer, and the modal interaction fusion is realized by the cross-attention module for the first time. Then, the features are spliced and input into the secondary Transformer for deep cross-modal fusion, and finally, the classification is completed by the multilayer perceptron. Experiments show that the CSTC model achieves overall classification accuracies of 92.32%, 99.81%, 97.90%, and 99.37% on the publicly available MUUFL dataset, Trento dataset, Augsburg dataset, and Houston2013 dataset, respectively, which is superior to the latest HSI–LiDAR separate classification algorithms. The ablation experiments and model performance evaluation experiments further show that the proposed CSTC model achieves excellent results in terms of robustness, adaptability, and parameter scale. Full article

Accurate nitrogen (N) content estimation in grassland vegetation is essential for ecosystem health and optimizing pasture quality, as N supports plant photosynthesis and water uptake. Traditional lab methods are slow and unsuitable for large-scale monitoring, while remote sensing models often face accuracy challenges due to hyperspectral data complexity. This study improves N content estimation in the typical steppe of Inner Mongolia by integrating hyperspectral remote sensing with advanced machine learning. Hyperspectral reflectance from Leymus chinensis and Cleistogenes squarrosa was measured using an ASD FieldSpec-4 spectrometer, and leaf N content was measured with an elemental analyzer. To address high-dimensional data, four spectral transformations—band combination, first-order derivative transformation (FDT), continuous wavelet transformation (CWT), and continuum removal transformation (CRT)—were applied, with Least Absolute Shrinkage and Selection Operator (LASSO) used for feature selection. Four machine learning models—Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Artificial Neural Network (ANN), and K-Nearest Neighbors (KNN)—were evaluated via five-fold cross-validation. Wavelet transformation provided the most informative parameters. The SVM model achieved the highest accuracy for L. chinensis (R² = 0.92), and the ANN model performed best for C. squarrosa (R² = 0.72). This study demonstrates that integrating wavelet transform with machine learning offers a reliable, scalable approach for grassland N monitoring and management. Full article

Computing power network (CPN) is designed to utilize multi-dimensional resources to complete computing tasks. However, in practical applications, the CPN architecture has difficulty in coordinating cross-domain heterogeneous resources, making it impossible to achieve the real-time and high scalability requirements of computationally intensive and time-sensitive tasks such as levee piping hazard inspection via remote sensing in emergency scenarios. Based on this, we propose a communication and computation integrated network architecture, referred to as (Com)²INet, that integrates “sensing”, “transmission”, and “computation” phases. In the sensing phase, thermal infrared imagery is utilized to retrieve land surface temperature fields through radiative transfer mechanisms, providing a reliable foundation for visual segmentation of piping hazards. In the transmission phase, we adopt the designed multi-path transmission mechanism to promote the efficient data flow across heterogeneous networks. In the computation phase, the proposed SACM algorithm, which is functionally decomposed and implemented as service chains within the proposed network architecture, dynamically processes the retrieved temperature fields to achieve precise hazard identification. This integrated framework ensures seamless interaction between sensing, communication, and computation, addressing the challenges of real-time hazard detection in emergency scenarios. Full article

This paper seeks to address the limitations of conventional remote sensing data dissemination algorithms, particularly their inability to model fine-grained multi-modal heterogeneous feature correlations and adapt to dynamic network topologies under resource constraints. This paper proposes multi-modal-MAPPO, a novel multi-modal deep reinforcement learning (MDRL) framework designed for a proactive data push in large-scale integrated satellite–terrestrial networks (ISTNs). By integrating satellite cache states, user cache states, and multi-modal data attributes (including imagery, metadata, and temporal request patterns) into a unified Markov decision process (MDP), our approach pioneers the application of the multi-actor-attention-critic with parameter sharing (MAPPO) algorithm to ISTNs push tasks. Central to this framework is a dual-branch actor network architecture that dynamically fuses heterogeneous modalities: a lightweight MobileNet-v3-small backbone extracts semantic features from remote sensing imagery, while parallel branches—a multi-layer perceptron (MLP) for static attributes (e.g., payload specifications, geolocation tags) and a long short-term memory (LSTM) network for temporal user cache patterns—jointly model contextual and historical dependencies. A dynamically weighted attention mechanism further adapts modality-specific contributions to enhance cross-modal correlation modeling in complex, time-varying scenarios. To mitigate the curse of dimensionality in high-dimensional action spaces, we introduce a multi-dimensional discretization strategy that decomposes decisions into hierarchical sub-policies, balancing computational efficiency and decision granularity. Comprehensive experiments against state-of-the-art baselines (MAPPO, MAAC) demonstrate that multi-modal-MAPPO reduces the average content delivery latency by 53.55% and 29.55%, respectively, while improving push hit rates by 0.1718 and 0.4248. These results establish the framework as a scalable and adaptive solution for real-time intelligent data services in next-generation ISTNs, addressing critical challenges in resource-constrained, dynamic satellite–terrestrial environments. Full article

Satellite remote sensing provides a cost-effective and large-scale alternative to traditional methods for retrieving water quality parameters for inland waters. Effective water quality parameter retrieval via optical satellite remote sensing requires three key components: (1) a sensor whose measurements are sensitive to variations in water quality; (2) accurate atmospheric correction to eliminate the effect of absorption and scattering in the atmosphere and retrieve the water-leaving radiance/reflectance; and (3) a bio-optical model used to estimate water quality from the optical signal. This study provides a literature review and an evaluation of these three components. First, a review of decommissioned, active, and upcoming satellite sensors is presented, highlighting their advantages and limitations, and a ranking method is introduced to assess their suitability for retrieving chlorophyll-a, colored dissolved organic matter, and non-algal particles in inland waters. This ranking can aid in selecting appropriate sensors for future studies. Second, the strengths and weaknesses of atmospheric correction algorithms used over inland waters are examined. The results show that no atmospheric correction algorithm performed consistently across all conditions. However, understanding their strengths and weaknesses allows users to select the most suitable algorithm for a specific use case. Third, the challenges, limitations, and recent advances of machine learning use in bio-optical models for inland water quality parameter retrieval are discussed. Machine learning models have limitations, including low generalizability, low dimensionality, spatial/temporal autocorrelation, and information leakage. These issues highlight the importance of locally trained models, rigorous cross-validation methods, and integrating auxiliary data to enhance dimensionality. Finally, recommendations for promising research directions are provided. Full article

Remote sensing is a valuable tool in precision agriculture due to its spatial and temporal coverage, non-destructive method of data collection, and cost-effectiveness. In this study, we measured the canopy reflectance of potato (Solanum tuberosum L.) crops on a plant-by-plant basis with a handheld spectrometer instrument. Our study pursues two primary objectives: (1) determining the optimal temporal aggregation for measuring canopy signals related to potato yield and (2) identifying the best spectral bands in the 350–2500 nm domain and vegetation indices. The study was conducted over two consecutive years (2020 and 2021) with 60 plants per plot, encompassing six potato varieties and three replicates annually throughout the growth season. Employing correlation analysis and dimensionality reduction, we identified 23 independent features significantly correlated with tuber yield. We used multiple linear regression analysis to model the relationship between the selected features and yield and to compare their influence in the fitted model. We used the Leave-One-Out Cross-Validation (LOOCV) method to assess the validity of the model (RMSE = 702 g and %RMSE = 29.2%). The most significant features included the Gitelson2 and Vogelmann indices. The optimal time period for measurements was determined to be from 56 to 100 days after planting. These findings may contribute to the advancement of precision farming by proposing tailored sensor applications, paving the way for improved agricultural practices and enhanced food security. Full article

Global dust events have become more frequent due to climate change and increased human activity, significantly impacting air quality and human health. Previous studies have mainly focused on determining atmospheric dust pollution levels through atmospheric parameter simulations or AOD values obtained from satellite remote sensing. However, research on the quantitative description of dust intensity and its cross-regional transport characteristics still faces numerous challenges. Therefore, this study utilized Fengyun-4A (FY-4A) satellite Advanced Geostationary Radiation Imager (AGRI) imagery, Cloud-Aerosol Lidar, and Infrared Pathfinder Satellite Observation (CALIPSO) lidar, and other auxiliary data, to conduct three-dimensional spatiotemporal monitoring and a cross-regional transport analysis of two typical dust events in the Beijing–Tianjin–Hebei (BTH) region of China using four dust intensity indices Infrared Channel Shortwave Dust (Icsd), Dust Detection Index (DDI), dust value (DV), and Dust Strength Index (DSI)) and the HYSPLIT model. We found that among the four indices, DDI was the most suitable for studying dust in the BTH region, with a detection accuracy (POCD) of >88% at all times and reaching a maximum of 96.14%. Both the 2021 and 2023 dust events originated from large-scale deforestation in southern Mongolia and the border area of Inner Mongolia, with dust plumes distributed between 2 and 12 km being transported across regions to the BTH area. Further, when dust aerosols are primarily concentrated below 4 km and PM10 concentrations consistently exceed 600 µg/m³, large dust storms are more likely to occur in the BTH region. The findings of this study provide valuable insights into the sources, transport pathways, and environmental impacts of dust aerosols. Full article