Search Results (6,582)

Cotton is a globally strategic crop that plays a crucial role in sustaining national economies and livelihoods. To address the challenges of accurate cotton field extraction in the complex planting environments of Xinjiang’s Alaer reclamation area, a cotton field identification model was developed that integrates multi-source satellite remote sensing data with machine learning methods. Using imagery from Sentinel-2, GF-1, and Landsat 8, we performed feature fusion using principal component, Gram–Schmidt (GS), and neural network techniques. Analyses of spectral, vegetation, and texture features revealed that the GS-fused blue bands of Sentinel-2 and Landsat 8 exhibited optimal performance, with a mean value of 16,725, a standard deviation of 2290, and an information entropy of 8.55. These metrics improved by 10,529, 168, and 0.28, respectively, compared with the original Landsat 8 data. In comparative classification experiments, the endmember-based random forest classifier (RFC) achieved the best traditional classification performance, with a kappa value of 0.963 and an overall accuracy (OA) of 97.22% based on 250 samples, resulting in a cotton-field extraction error of 38.58 km². By enhancing the deep learning model, we proposed a U-Net architecture that incorporated a Convolutional Block Attention Module and Atrous Spatial Pyramid Pooling. Using the GS-fused blue band data, the model achieved significantly improved accuracy, with a kappa coefficient of 0.988 and an OA of 98.56%. This advancement reduced the area estimation error to 25.42 km², representing a 34.1% decrease compared with that of the RFC. Based on the optimal model, we constructed a digital map of continuous cotton cropping from 2021 to 2023, which revealed a consistent decline in cotton acreage within the reclaimed areas. This finding underscores the effectiveness of crop rotation policies in mitigating the adverse effects of large-scale monoculture practices. This study confirms that the synergistic integration of multi-source satellite feature fusion and deep learning significantly improves crop identification accuracy, providing reliable technical support for agricultural policy formulation and sustainable farmland management. Full article

Space-based on-orbit servicing missions impose stringent requirements for precise identification and localization of satellite components, while existing detection algorithms face dual challenges of insufficient accuracy and excessive computational resource consumption. This paper proposes SLD-YOLO, a lightweight satellite component detection model based on improved YOLO11, balancing accuracy and efficiency through structural optimization and lightweight design. First, we design RLNet, a lightweight backbone network that employs reparameterization mechanisms and hierarchical feature fusion strategies to reduce model complexity by 19.72% while maintaining detection accuracy. Second, we propose the CSP-HSF multi-scale feature fusion module, used in conjunction with PSConv downsampling, to effectively improve the model’s perception of multi-scale objects. Finally, we introduce SimAM, a parameter-free attention mechanism in the detection head to further improve feature representation capability. Experiments on the UESD dataset demonstrate that SLD-YOLO achieves measurable improvements compared to the baseline YOLO11s model across five satellite component detection categories: mAP50 increases by 2.22% to 87.44%, mAP50:95 improves by 1.72% to 63.25%, while computational complexity decreases by 19.72%, parameter count reduces by 25.93%, model file size compresses by 24.59%, and inference speed reaches 90.4 FPS. Validation experiments on the UESD_edition2 dataset further confirm the model’s robustness. This research provides an effective solution for target detection tasks in resource-constrained space environments, demonstrating practical engineering application value. Full article

Proxy data-based reconstructions provide an essential basis for understanding comprehensive precipitation variability at multiple time scales. This study compared the variation characteristics of reconstructed precipitation data across different regions in the U.S. and the differences at decadal/multidecadal scales. The reconstruction showed that multiple scales of precipitation variability existed in each region and both multidecadal and decadal variability varied over time and across region. There was weaker multidecadal variability in the latter half of the 18th century and during the mid-19th century to mid-20th century east of the Rocky Mountains (RM); however, multidecadal variability appears to have increased since the 20th century in most regions. Decadal variability was weaker west of the RM except in the Southwest U.S. in the latter half of the 18th century. While decadal variability became stronger in the early 20th century, it shifted from a stronger phase to a weaker phase east of the RM. Then, we compared the spatiotemporal differences between the reconstructed Palmer Drought Severity Index (PDSI) and reconstructed precipitation in this study. The reconstructed annual precipitation mostly remains consistent with the existing PDSI dataset, but there are inconsistencies in the severe dry/wet intensities in some regions. Multiscale analysis of regional precipitation data holds great importance for understanding the relationship between precipitation in different regions and the climate system, while also providing a scientific theoretical basis for precipitation prediction. Full article

With the rapid expansion of multi-platform remote sensing applications, cloud contamination significantly impedes cross-platform data utilization. Current cloud detection methods face critical technical challenges in cross-platform settings, including neglect of atmospheric radiative transfer mechanisms, inadequate multi-scale structural decoupling, high intra-class variability coupled with inter-class similarity, cloud boundary ambiguity, cross-modal feature inconsistency, and noise propagation in pseudo-labels within semi-supervised frameworks. To address these issues, we introduce a Physics-Guided Multi-Representation Network (PGMRN) that adopts a student–teacher architecture and fuses tri-modal representations—Pseudo-NDVI, structural, and textural features—via atmospheric priors and intrinsic image decomposition. Specifically, PGMRN first incorporates an InfoNCE contrastive loss to enhance intra-class compactness and inter-class discrimination while preserving physical consistency; subsequently, a boundary-aware regional adaptive weighted cross-entropy loss integrates PA-CAM confidence with distance transforms to refine edge accuracy; furthermore, an Uncertainty-Aware Quadruple Consistency Propagation (UAQCP) enforces alignment across structural, textural, RGB, and physical modalities; and finally, a dynamic confidence-screening mechanism that couples PA-CAM with information entropy and percentile-based thresholding robustly refines pseudo-labels. Extensive experiments on four benchmark datasets demonstrate that PGMRN achieves state-of-the-art performance, with Mean IoU values of 70.8% on TCDD, 79.0% on HRC_WHU, and 83.8% on SWIMSEG, outperforming existing methods. Full article

Global ambitions for net-zero emissions and resource circularity are propelling industry from linear “make-use-dispose”models toward closed-loop value creation. Remanufacturing, which aims to restore end-of-life products to a “like-new” condition, plays a central role in this transition. However, its stochastic inputs and complex, multi-stage processes pose significant challenges to traditional production planning methods. This study delivers an integrated overview of remanufacturing scheduling by combining a systematic bibliometric review of 190 publications (2005–2025) with a critical synthesis of modelling approaches and enabling technologies. The bibliometric results reveal five thematic clusters and a 14% annual growth rate, highlighting a shift from deterministic, shop-floor-focused models to uncertainty-aware, sustainability-oriented frameworks. The scheduling problems are formalised to capture features arising from variable core quality, multi-phase precedence, and carbon reduction goals, in both centralised and cloud-based systems. Advances in human–robot disassembly, vision-based inspection, hybrid repair, and digital testing demonstrate feedback-rich environments that increasingly integrate planning and execution. A comparative analysis shows that, while mixed-integer programming and metaheuristics perform well in small static settings, dynamic and large-scale contexts benefit from reinforcement learning and hybrid decomposition models. Finally, future directions for dynamic, collaborative, carbon-conscious, and digital-twin-driven scheduling are outlined and investigated. Full article

The precise mapping of rock joint traces is fundamental to the design and safety assessment of foundations, retaining structures, and underground cavities in building and civil engineering. Existing deep learning approaches either impose prohibitive computational demands for on-site deployment or disrupt the topological continuity of subpixel lineaments that govern rock mass behavior. This study presents BATNet-Lite, a lightweight encoder–decoder architecture optimized for joint segmentation on resource-constrained devices. The encoder introduces a Boundary-Aware Token-Mixing (BATM) block that separates feature maps into patch tokens and directionally pooled stripe tokens, and a bidirectional attention mechanism subsequently transfers global context to local descriptors while refining stripe features, thereby capturing long-range connectivity with negligible overhead. A complementary Multi-Scale Line Enhancement (MLE) module combines depth-wise dilated and deformable convolutions to yield scale-invariant responses to joints of varying apertures. In the decoder, a Skeletal-Contrastive Decoder (SCD) employs dual heads to predict segmentation and skeleton maps simultaneously, while an InfoNCE-based contrastive loss enforces their topological consistency without requiring explicit skeleton labels. Training leverages a composite focal Tversky and edge IoU loss under a curriculum-thinning schedule, improving edge adherence and continuity. Ablation experiments confirm that BATM, MLE, and SCD each contribute substantial gains in boundary accuracy and connectivity preservation. By delivering topology-preserving joint maps with small parameters, BATNet-Lite facilitates rapid geological data acquisition for tunnel face mapping, slope inspection, and subsurface digital twin development, thereby supporting safer and more efficient building and underground engineering practice. Full article

Mountain glaciers are the most direct and sensitive indicators of climate change. In the context of global warming, monitoring changes in glacier elevation has become a crucial issue in modern cryosphere research. The Global Ecosystem Dynamics Investigation (GEDI) is a full-waveform laser altimeter with a multi-beam that provides unprecedented measurements of the Earth’s surface. Many studies have investigated its applications in assessing the vertical structure of various forests. However, few studies have assessed GEDI’s performance in detecting variations in glacier elevation in land ice in high-mountain Asia. To address this limitation, we selected the Southeastern Tibetan Plateau (SETP), one of the most sensitive areas to climate change, as a test area to assess the feasibility of using GEDI to monitor glacier elevation changes by comparing it with ICESat-2 ATL06 and the reference TanDEM-X DEM products. Moreover, this study further analyzes the influence of environmental factors (e.g., terrain slope and aspect, and altitude distribution) and glacier attributes (e.g., glacier area and debris cover) on changes in glacier elevation. The results show the following: (1) Compared to ICESat-2, in most cases, GEDI overestimated glacier thinning (i.e., elevation reduction) to some extent from 2019 to 2021, with an average overestimation value of about −0.29 m, while the annual average rate of elevation change was relatively close, at −0.70 ± 0.12 m/yr versus −0.62 ± 0.08 m/yr, respectively. (2) In terms of time, GEDI reflected glacier elevation changes at interannual and seasonal scales, and the trend of change was consistent with that found with ICESat-2. The results indicate that glacier accumulation mainly occurred in spring and winter, while the melting rate accelerated in summer and autumn. (3) GEDI effectively monitored and revealed the characteristics and patterns of glacier elevation changes with different terrain features, glacier area grades, etc.; however, as the slope increased, the accuracy of the reported changes in glacier elevation gradually decreased. Nonetheless, GEDI still provided reasonable estimates for changes in mountain glacier elevation. (4) The spatial distribution of GEDI footprints was uneven, directly affecting the accuracy of the monitoring results. Thus, to improve analyses of changes in glacier elevation, terrain factors should be comprehensively considered in further research. Overall, these promising results have the potential to be used as a basic dataset for further investigations of glacier mass and global climate change research. Full article

Pipelines play a vital role in material transportation within industrial settings. This review synthesizes detection technologies for early-stage small gas leaks from pipelines in the industrial sector, with a focus on acoustic-based methods, optical gas imaging (OGI), and multimodal fusion approaches. It encompasses detection principles, inherent challenges, mitigation strategies, and the state of the art (SOTA). Small leaks refer to low flow leakage originating from defects with apertures at millimeter or submillimeter scales, posing significant detection difficulties. Acoustic detection leverages the acoustic wave signals generated by gas leaks for non-contact monitoring, offering advantages such as rapid response and broad coverage. However, its susceptibility to environmental noise interference often triggers false alarms. This limitation can be mitigated through time-frequency analysis, multi-sensor fusion, and deep-learning algorithms—effectively enhancing leak signals, suppressing background noise, and thereby improving the system’s detection robustness and accuracy. OGI utilizes infrared imaging technology to visualize leakage gas and is applicable to the detection of various polar gases. Its primary limitations include low image resolution, low contrast, and interference from complex backgrounds. Mitigation techniques involve background subtraction, optical flow estimation, fully convolutional neural networks (FCNNs), and vision transformers (ViTs), which enhance image contrast and extract multi-scale features to boost detection precision. Multimodal fusion technology integrates data from diverse sensors, such as acoustic and optical devices. Key challenges lie in achieving spatiotemporal synchronization across multiple sensors and effectively fusing heterogeneous data streams. Current methodologies primarily utilize decision-level fusion and feature-level fusion techniques. Decision-level fusion offers high flexibility and ease of implementation but lacks inter-feature interaction; it is less effective than feature-level fusion when correlations exist between heterogeneous features. Feature-level fusion amalgamates data from different modalities during the feature extraction phase, generating a unified cross-modal representation that effectively resolves inter-modal heterogeneity. In conclusion, we posit that multimodal fusion holds significant potential for further enhancing detection accuracy beyond the capabilities of existing single-modality technologies and is poised to become a major focus of future research in this domain. Full article

With the accelerating global energy transition, wind power has become a core pillar of renewable energy systems. However, its inherent intermittency and volatility pose significant challenges to the safe, stable, and economical operation of power grids—making ultra-short-term wind power prediction a critical technical link in optimizing grid scheduling and promoting large-scale wind power integration. Current forecasting techniques are plagued by problems like the inadequate representation of features, the poor separation of features, and the challenging clarity of deep learning models. This study introduces a method for the prediction of wind energy using spatiotemporal contrastive learning, employing seasonal trend decomposition to encapsulate the diverse characteristics of time series. A contrastive learning framework and a feature disentanglement loss function are employed to effectively decouple spatiotemporal features. Data on geographical positions are integrated to simulate spatial correlations, and a convolutional network of spatiotemporal graphs, integrated with a multi-head attention system, is crafted to improve the clarity. The proposed method is validated using operational data from two actual wind farms in Northwestern China. The research indicates that, compared with typical baselines (e.g., STGCN), this method reduces the RMSE by up to 38.47% and the MAE by up to 44.71% for ultra-short-term wind power prediction, markedly enhancing the prediction precision and offering a more efficient way to forecast wind power. Full article

The acquisition of nighttime remote-sensing visible-light images is often accompanied by low-illumination effects and haze interference, resulting in significant image quality degradation and greatly affecting subsequent applications. Existing low-light enhancement and dehazing algorithms can handle each problem individually, but their simple cascade cannot effectively address unknown real-world degradations. Therefore, we design a joint processing framework, WFDiff, which fully exploits the advantages of Fourier–wavelet dual-domain features and innovatively integrates the inverse diffusion process through differentiable operators to construct a multi-scale degradation collaborative correction system. Specifically, in the reverse diffusion process, a dual-domain feature interaction module is designed, and the joint probability distribution of the generated image and real data is constrained through differentiable operators: on the one hand, a global frequency-domain prior is established by jointly constraining Fourier amplitude and phase, effectively maintaining the radiometric consistency of the image; on the other hand, wavelets are used to capture high-frequency details and edge structures in the spatial domain to improve the prediction process. On this basis, a cross-overlapping-block adaptive smoothing estimation algorithm is proposed, which achieves dynamic fusion of multi-scale features through a differentiable weighting strategy, effectively solving the problem of restoring images of different sizes and avoiding local inconsistencies. In view of the current lack of remote-sensing data for low-light haze scenarios, we constructed the Hazy-Dark dataset. Physical experiments and ablation experiments show that the proposed method outperforms existing single-task or simple cascade methods in terms of image fidelity, detail recovery capability, and visual naturalness, providing a new paradigm for remote-sensing image processing under coupled degradations. Full article

Optical coherence tomography (OCT) acquisitions often reduce lateral sampling density to shorten scan time and suppress motion artifacts, but this strategy degrades the signal-to-noise ratio and obscures fine retinal microstructures. To recover these details without hardware modifications, we propose MGFormer, a lightweight Transformer for OCT super-resolution (SR) that integrates a multi-granularity attention mechanism with tensor distillation. A feature-enhancing convolution first sharpens edges; stacked multi-granularity attention blocks then fuse coarse-to-fine context, while a row-wise top-k operator retains the most informative tokens and preserves their positional order. We trained and evaluated MGFormer on B-scans from the Duke SD-OCT dataset at $2\times$, $4\times$, and $8\times$ scaling factors. Relative to seven recent CNN- and Transformer-based SR models, MGFormer achieves the highest quantitative fidelity; at $4\times$ it reaches 34.39 dB PSNR and 0.8399 SSIM, surpassing SwinIR by +0.52 dB and +0.026 SSIM, and reduces LPIPS by 21.4%. Compared with the same backbone without tensor distillation, FLOPs drop from 289G to 233G (−19.4%), and per-B-scan latency at $4\times$ falls from 166.43 ms to 98.17 ms (−41.01%); the model size remains compact (105.68 MB). A blinded reader study shows higher scores for boundary sharpness (4.2 ± 0.3), pathology discernibility (4.1 ± 0.3), and diagnostic confidence (4.3 ± 0.2), exceeding SwinIR by 0.3–0.5 points. These results suggest that MGFormer can provide fast, high-fidelity OCT SR suitable for routine clinical workflows. Full article

Traditional information distillation networks using single-scale convolution and simple feature fusion often result in insufficient information extraction and ineffective restoration of high-frequency details. To address this problem, we propose a lightweight image super-resolution reconstruction network based on multi-order information optimization. The core of this network lies in the enhancement and refinement of high-frequency information. Our method operates through two main stages to fully exploit the high-frequency features in images while eliminating redundant information, thereby enhancing the network’s detail restoration capability. In the high-frequency information enhancement stage, we design a self-calibration high-frequency information enhancement block. This block generates calibration weights through self-calibration branches to modulate the response strength of each pixel. It then selectively enhances critical high-frequency information. Additionally, we combine an auxiliary branch and a chunked space optimization strategy to extract local details and adaptively reinforce high-frequency features. In the high-frequency information refinement stage, we propose a multi-scale high-frequency information refinement block. First, multi-scale information is captured through multiplicity sampling to enrich the feature hierarchy. Second, the high-frequency information is further refined using a multi-branch structure incorporating wavelet convolution and band convolution, enabling the extraction of diverse detailed features. Experimental results demonstrate that our network achieves an optimal balance between complexity and performance, outperforming popular lightweight networks in both quantitative metrics and visual quality. Full article

Background: Pan-drug-resistant (PDR) Acinetobacterinfections are an escalating ICU threat, demanding both patient-level triage and facility-wide forecasting. Objective: The aim of this study was to build a dual-scale AI framework that (i) predicts PDR status at infection onset and (ii) forecasts hospital-level PDR burden through 2027. Methods: We retrospectively analyzed 270 Acinetobacter infection episodes (2020–2024) with 65 predictors spanning demographics, timelines, infection type, resistance-class flags, and a 25-drug antibiogram. TabTransformer and XGBoost were trained on 2020–2023 episodes (n = 210), evaluated by stratified 5-fold CV, and externally tested on 2024 episodes (n = 60). Metrics included AUROC, AUPRC, accuracy, and recall at 90% specificity; AUROC was optimism-corrected via 0.632 + bootstrap and DeLong-tested for drift. SHAP values quantified feature impact. Weekly PDR incidence was forecast with an attention–LSTM model retrained monthly (200 weekly origins, 4-week horizon) and benchmarked against seasonal-naïve, Prophet, and SARIMA models (MAPE and RMSE). Quarterly projections (TFT-lite) extended forecasts to 2027. Results: The CV AUROC was 0.924 (optimism-corrected 0.874); an ensemble of TabTransformer + XGBoost reached 0.958. The 2024 AUROC fell to 0.586 (p < 0.001), coinciding with a PDR prevalence drop (75→38%) and three covariates with PSIs > 1.0. Isotonic recalibration improved the Brier score from 0.326 to 0.207 and yielded a net benefit equivalent to 26 unnecessary isolation-days averted per 100 ICU admissions at a 0.20 threshold. SHAP highlighted Ampicillin/Sulbactam resistance, unknown acquisition mode, and device-related infection as dominant drivers. The attention–LSTM achieved a median weekly MAE of 0.10 (IQR: 0.028–0.985) vs. 1.00 for the seasonal-naïve rule, outperforming it on 48.5% of weeks and surpassing Prophet and SARIMA (MAPE = 6.2%, RMSE = 0.032). TFT-lite projected a ≥ 25% PDR tipping point in 2025 Q1 with a sustained rise in 2027. Conclusions: The proposed framework delivers explainable patient-level PDR risk scores and competitive 4-week and multi-year incidence forecasts despite temporal drift, supporting antimicrobial stewardship and ICU capacity planning. Shrinkage and bootstrap correction were applied to address the small sample size (EPV = 2.1), which poses an overfitting risk. Continuous recalibration and multi-center validation remain priorities. Full article

Gaze estimation is a cornerstone of applications such as human–computer interaction and behavioral analysis, e.g., for intelligent transport systems. Nevertheless, existing methods predominantly rely on coarse-grained features from deep layers of visual encoders, overlooking the critical role that fine-grained details from shallow layers play in gaze estimation. To address this gap, we propose a novel Hierarchical Fine-Grained Attention Decoder (HFGAD), a lightweight fine-grained decoder that emphasizes the importance of shallow-layer information in gaze estimation. Specifically, HFGAD integrates a fine-grained amplifier MSCSA that employs multi-scale spatial-channel attention to direct focus toward gaze-relevant regions, and also incorporates a shallow-to-deep fusion module SFM to facilitate interaction between coarse-grained and fine-grained information. Extensive experiments on three benchmark datasets demonstrate the superiority of HFGAD over existing methods, achieving a remarkable 1.13° improvement in gaze estimation accuracy for in-car scenarios. Full article

Diabetic retinopathy (DR), a critical ocular disease that can lead to blindness, demands early and accurate diagnosis to prevent vision loss. Current automated DR diagnosis methods face two core challenges: first, subtle early lesions such as microaneurysms are often missed due to insufficient feature extraction; second, there is a persistent trade-off between model accuracy and efficiency—lightweight architectures often sacrifice precision for real-time performance, while high-accuracy models are computationally expensive and difficult to deploy on resource-constrained edge devices. To address these issues, this study presents a novel deep learning framework integrating depthwise separable convolution and a multi-view attention mechanism (MVAM) for efficient DR diagnosis using retinal images. The framework employs multi-scale feature fusion via parallel 3 × 3 and 5 × 5 convolutions to capture lesions of varying sizes and incorporates Gabor filters to enhance vascular texture and directional lesion modeling, improving sensitivity to early structural abnormalities while reducing computational costs. Experimental results on both the diabetic retinopathy (DR) dataset and ocular disease (OD) dataset demonstrate the superiority of the proposed method: it achieves a high accuracy of 0.9697 on the DR dataset and 0.9669 on the OD dataset, outperforming traditional methods such as CNN_eye, VGG, and UNet by more than 1 percentage point. Moreover, its training time is only half that of U-Net (on DR dataset) and VGG (on OD dataset), highlighting its potential for clinical DR screening. Full article