Search Results (54)

Accurate monitoring of hydrological dynamics in complex perennial landscapes is a cornerstone for tropical agricultural sustainability. Traditional energy balance models based on orbital optical data often face methodological bottlenecks due to cloud cover and the “greening myth,” where optical indices fail to capture immediate water stress due to the non-linear decoupling between stomatal closure and pigment loss. This study developed a cloud-integrated multisensor framework to estimate actual evapotranspiration (ET_a) at a refined 100 m resolution in mountain coffee systems, utilizing active microwave proxies from Sentinel-1. We fused polarimetric metrics—Degree of Polarization (DoP) and Shannon Entropy (SE)—with land surface temperature and soil moisture data. Multiple Linear Regression (MLR) was compared against non-linear algorithms (Random Forest and SVR) to prioritize model parsimony and physical interpretability. The results show that MLR emerged as the most parsimonious and suitable model within this localized dataset scope (R² = 0.872; RMSE = 2.916 mm/8-day), outperforming complex “black-box” architectures. Soil moisture emerged as the dominant environmental driver of ET_a variability, while SAR-based metrics served as sensitive mechanical proxies for canopy geometric heterogeneity and macro-structural variations. Cross-correlation analysis revealed a 16-day lag, empirically indicating that biophysical water shifts temporally precede geometric canopy alterations. Operationally, this framework ensures temporal continuity under persistent cloud cover and provides high-fidelity spatial detailing for precision water management. This approach offers an auditable and scalable tool for watershed planning and climate resilience in tropical agriculture. Full article

The social–ecological nexus (SEN) offers a framework to capture the complex and dynamic interactions and interdependencies between human communities and the natural systems that support them. This study analyzed the SENs within a village tank cascade system (VTCS), a social–ecological system (SES) located in the dry zone of Sri Lanka. The study adopted a participatory approach, combining fuzzy cognitive mapping (FCM) to determine key SES properties of the VTCS. The FCM process identified 49 nodes (elements) and 434 edges (connections) within the study landscape that contribute to system performance. Network graphs were generated using centrality metrics—degree, betweenness, and eigenvector centrality—to identify the most influential nodes and edges contributing to system sustainability and productivity. The study identified nine nodes as the most influential elements in the SEN which are critical for balancing trade-offs between sustainability and productivity in the VTCS. Three distinct clusters of elements influencing sustainability and productivity emerged from the SEN graph: (i) ecological cluster, (ii) social–ecological cluster, and (iii) social cluster. Understanding the role of SES elements and their positions in the SEN is crucial for identifying gaps within the system and informing tailored management interventions. These findings offer a theoretical basis for optimizing sustainability strategies aimed at enhancing the overall productivity and resilience of SES. Consequently, this approach exposes the complexities of the SEN, making it widely applicable to similar SESs globally. Full article

Reliable precipitation projections are essential for water-resource management, flood-risk assessment, and drought preparedness in hydroclimatically complex regions such as Mexico, where uncertainty remains high due to monsoon dynamics, complex topography, and tropical moisture transport. This study evaluates paired CMIP5 and CMIP6 global climate models in simulating the historical (1940–2005) precipitation annual cycle across four regions of Mexico (NW, NE, SW, SE). Model outputs were compared against ERA5 and cross-validated with CRU using complementary metrics assessing error magnitude, variability, temporal phase, and spatial coherence. Results indicate that CMIP6 provides moderate but regionally heterogeneous improvements rather than a uniform advance. The most consistent gains occur in NE and SE Mexico, where dry biases are reduced and seasonal amplitude is better represented. In contrast, SW Mexico exhibits persistent summer wet biases linked to monsoon–topography interactions, while improvements in NW Mexico are mainly confined to selected individual CMIP6 models and are not consistently reflected in the ensemble median. A marked SW–SE summer dipole bias highlights ongoing deficiencies in representing moisture transport and convection. These findings demonstrate that increased model complexity does not guarantee improved regional skill and that ensemble medians may mask individual model performance, underscoring the need for targeted model selection, multi-dataset validation, and bias-correction strategies. Full article

Transformer-based speech enhancement (SE) architectures suffer from high computational complexity, while existing lightweight state space model (SSM) approaches are constrained to fixed one-dimensional scanning that cannot fully exploit the two-dimensional time–frequency structure of speech spectrograms. To address these limitations, we propose DOM-MUSE, a lightweight U-Net-style SE framework built upon the Mamba-2 SSM with four targeted innovations. First, a Deformable Feature Extractor (DFE) predicts per location spatial offsets that warp the feature sampling grid to align with speech formant trajectories and harmonic structures, providing geometrically coherent inputs to the state space model. Second, a DOM Mamba Block with Cross-Dimensional Gated Fusion (CDGF) deploys two parallel Mamba-2 instances scanning the time and frequency axes independently, and uses Taylor Channel Attention (TCA) to derive semantic gates that modulate each SSM output before fusion. Third, a Phase-Guided Feature Conditioner (PGFC) computes local phase-gradient gates that suppress noise-dominated activations prior to the SSM stage, making the feature extraction pathway implicitly phase-aware. Fourth, an Attention-Based Skip Connection (ABSC) replaces conventional concatenation skip connections with a learned channel gate, adaptively controlling the information flow from the encoder to the decoder. Experiments on the VoiceBank-DEMAND benchmark demonstrate that DOM-MUSE outperforms the reproduced MUSE baseline on all five evaluation metrics—including PESQ (+0.077), CSIG (+0.058), CBAK (+0.026), COVL (+0.070), and STOI (+0.002)—while reducing the parameter count by 24% (0.51 M to 0.39 M). Notably, DOM-MUSE also surpasses MUSE++ on perceptual quality metrics (PESQ +0.061, COVL +0.032) despite MUSE++ employing dynamic SNR augmentation and an augmented multi-objective loss that DOM-MUSE deliberately omits, demonstrating that the proposed architectural innovations yield genuine improvements independent of training strategy. When DOM-MUSE is additionally trained under the same augmented protocol as MUSE++, it achieves PESQ of 3.46 and COVL of 4.22, further confirming the complementary nature of architectural and training improvements. Full article

As software systems continue to grow in scale and complexity, fuzzing has become an indispensable automated technique for vulnerability discovery. Compared with coverage-guided fuzzing, directed greybox fuzzing (DGF) focuses execution toward specific basic blocks or functions, making it widely used in scenarios such as patch testing and vulnerability reproduction. Recent studies have combined fuzzing with symbolic execution (SE) to generate inputs that are difficult to obtain through mutation alone. However, applying SE to all branch conditions along an execution path may explore many paths unrelated to the target, leading to substantial overhead in directed fuzzing. Meanwhile, existing distance metrics still have limitations in guiding seeds toward targets: AFLGo relies on structural control-flow distances, which may not precisely reflect target reachability, while existing probability-based metrics often simplify complex control-flow structures such as loops and back-edges. To address these limitations, we propose TriFuzz, a probabilistic distance-guided hybrid directed fuzzing framework that integrates a loop-aware reachability distance model, target-related selective symbolic instrumentation, and a tightly coupled AFLGo–SymCC coordination mechanism. TriFuzz uses the probability-based distance model as the primary guidance signal and applies selective symbolic instrumentation to prune irrelevant basic blocks and concentrate exploration on target-relevant code regions. Our evaluation on the AFLGo testsuite and UniBench shows that TriFuzz improves both time-to-target and time-to-exposure on most evaluated benchmarks, demonstrating the effectiveness of combining fine-grained probabilistic distance guidance with selective symbolic reasoning and tightly integrated hybrid execution. Full article

Knowledge graphs play a vital role in tasks such as recommendation systems, question-answering systems, and information retrieval. However, during practical construction, they commonly suffer from structural incompleteness and sparse relationships, which limit reasoning performance and downstream applications. Existing methods typically focus solely on either structural modeling or semantic modeling: embedding models relying solely on graph structures struggle to leverage textual information about entities and relationships. In contrast, semantic approaches relying solely on pre-trained language models struggle to accurately capture complex relationship patterns. To address this challenge, this paper proposes SeSKGC, a semantic–structural fusion knowledge graph completion model. At the semantic level, the model employs the DeBERTa pre-trained language model to encode entity and relation text. It incorporates a neighbor text augmentation mechanism to introduce local semantic context and enhance representation quality. At the structural level, it adopts complex-space rotation to model relationships using a RotatE-like approach, and aggregates local topological information through relative position attention to capture complex relationship patterns. At the scoring stage, the model employs a weighted fusion strategy to combine semantic and structural scores and utilizes InfoNCE contrastive loss for joint optimization. Experiments conducted on WN18RR and FB15k-237 datasets demonstrate that SeSKGC achieves overall superior performance on metrics including MRR and Hits@N compared to multiple representative baseline methods. Ablation studies and parameter sensitivity analysis of fusion weight $\lambda$ further reveal that the semantic encoding and structural modeling modules exhibit distinct complementary roles, while the weighted fusion design in the scoring layer plays a crucial role in enhancing model performance and stability. Full article

Intentional high-power electromagnetic (EM) interference poses a serious threat to sensitive electronic systems and often manifests as ultra-wideband (UWB) sub- and nanosecond pulses. Metallic shielding enclosures with technological apertures are commonly used for protection; however, apertures enable electromagnetic coupling into the enclosure and limit shielding performance. While most existing studies focus on transient disturbances with durations exceeding the enclosure transit time, this work addresses an ultrashort high-power subnanosecond UWB plane-wave pulse whose duration is significantly shorter than the enclosure transit time, a regime that remains insufficiently explored. A time-domain numerical analysis is performed for a low-profile rectangular metallic enclosure with a front-wall aperture, focusing on internal EM field evolution, internal pulse formation, and polarization-dependent shielding effectiveness. Three-dimensional full-wave simulations were carried out using CST Microwave Studio over a 90 ns observation window. The results show that the incident pulse excites primary subnanosecond EM waves inside the enclosure, which subsequently generate secondary waves through multiple reflections from the enclosure walls. Their interaction produces complex, long-lasting, time-varying internal field patterns. Although attenuated, the resulting internal subnanosecond pulses repeatedly traverse the enclosure interior, forming a pulse train-like sequence that may pose a cumulative electromagnetic threat to internal electronics. A key contribution of this work is the quantification of time-dependent local shielding effectiveness for both electric and magnetic fields, derived directly from the internal pulse train-like series obtained in the time domain. The concept of local, time-dependent shielding effectiveness provides physical insight that cannot be obtained from a single globally averaged SE value. In the case of ultrashort electromagnetic pulse excitation, the internal field response of an enclosure is strongly non-stationary and highly non-uniform in space, with local field maxima occurring at specific times and locations despite good average shielding performance. Time-dependent local SE enables identification of worst-case temporal conditions, repeated high-amplitude internal exposures, and critical regions inside the enclosure where shielding is significantly weaker than suggested by global metrics. Therefore, while conventional SE remains useful as a summary measurand, local time-dependent SE is essential for assessing the actual electromagnetic risk to sensitive electronics under ultrashort pulse disturbances. In addition, a global shielding effectiveness metric mapped over selected enclosure cross-sections is introduced to enable rapid visual assessment of shielding performance. The analysis demonstrates a strong dependence of internal wave propagation, internal pulse formation, and both local and global shielding effectiveness on the polarization of the incident subnanosecond EM pulse. These findings provide new physical insight into aperture coupling and shielding behavior in the ultrashort-pulse regime and offer practical guidance for the assessment and design of compact shielding enclosures exposed to high-power UWB EM threats. Full article

Montane forests, characterized by complex terrain and diverse climates, serve as critical global biodiversity hotspots, particularly for birds and mammals. However, the patterns and underlying processes of bird and mammal diversity remain insufficiently studied in the montane forests of eastern China. This study employed infrared-triggered camera trapping to conduct a four-year field monitoring of birds and mammals, analyzing the effects of plant diversity and seasonal variations on the diversity of habitat-associated animals. Our results revealed that species-level habitat visit frequency in ground-dwelling birds exhibited a significant phylogenetic signal, particularly in spring and summer. Plant diversity metrics demonstrated significant positive correlations with corresponding bird metrics of species richness (SR), phylogenetic diversity (PD), and the standardized effect size of PD (Phylo SES PD). In contrast, for mammals, plant diversity metrics were significantly positively correlated with corresponding SR, mean pairwise phylogenetic distance (Phylo MPD), and mean nearest phylogenetic taxon distance (Phylo MNTD), as well as community structure metrics, including the net relatedness index (Phylo NRI) and nearest taxon index (Phylo NTI). Furthermore, the plant Shannon–Wiener index showed significant positive correlations with both bird and mammal metrics of SR, PD, and Phylo SES PD but significant negative correlations with Phylo MNTD. Seasonal variations triggered the mean altitudinal migration in ground-dwelling birds and mammals. There were significant differences in the diversity and community structure metrics of birds (Shannon–Wiener, Funct FNND, and PD) and mammals (Shannon–Wiener, Funct MPD, Funct FNND, PD, Phylo MPD, Phylo MNTD, and Phylo SES PD), which varied across different seasons. These findings emphasize that plant diversity and seasonal changes are closely related to the diversity and community structure of birds and mammals. They provide theoretical support for the role of habitat vegetation and seasonal dynamics in maintaining the stability and functioning of montane animal ecosystems, offering important insights for addressing habitat fragmentation and species migratory behavior. Full article

Quantitative precipitation estimation (QPE) from radar reflectivity is fundamental for weather nowcasting and water resource management. Conventional Z-R relationship formulas, derived from Rayleigh scattering theory, rely heavily on empirical parameter fitting, which limits the estimation accuracy and generalization across different precipitation regimes. Recent deep learning (DL)-based QPE methods can capture the complex nonlinear relationships between radar reflectivity and rainfall. However, most of them overlook fundamental physical constraints, resulting in reduced robustness and interpretability. To address these issues, this paper proposes FusionQPE, a novel Physics-Constrained DL framework that integrates an adaptive Z-R formula. Specifically, FusionQPE employs a Dense convolutional neural network (DenseNet) backbone to extract multi-scale spatial features from radar echoes, while a modified squeeze-and-excitation (SE) network adaptively learns the parameters of the Z-R relationship. The final rainfall estimate is obtained through a linear combination of outputs from both the DenseNet backbone and the adaptive Z-R branch, where the trained linear weight and Z-R parameters provide interpretable insights into the model’s physical reasoning. Moreover, a physical-based constraint derived from the Z-R branch output is incorporated into the loss function to further strengthen physical consistency. Comprehensive experiments on real radar and rain gauge observations from Guangzhou, China, demonstrate that FusionQPE consistently outperforms both traditional and state-of-the-art DL-based QPE models across multiple evaluation metrics. The ablation and interpretability analysis further confirms that the adaptive Z-R branch improves both the physical consistency and credibility of the model’s precipitation estimation. Full article

Deep learning has become a powerful tool for diagnosing pests and plant diseases, although conventional convolutional neural networks (CNNs) generally suffer from limited interpretability and suboptimal focus on important image features. This study examines the integration of attention mechanisms into two prevalent CNN architectures—ResNet50 and MobileNetV2—to improve the interpretability and classification of diseases impacting date palm trees. Four attention modules—Squeeze-and-Excitation (SE), Efficient Channel Attention (ECA), Soft Attention, and the Convolutional Block Attention Module (CBAM)—were systematically integrated into ResNet50 and MobileNetV2 and assessed on the Palm Leaves dataset. Using transfer learning, the models were trained and evaluated through accuracy, F1-score, Grad-CAM visualizations, and quantitative metrics such as entropy and Attention Focus Scores. Analysis was also performed on the model’s complexity, including parameters and FLOPs. To confirm generalization, we tested the improved models on field data that was not part of the dataset used for learning. The experimental results demonstrated that the integration of attention mechanisms substantially improved both predictive accuracy and interpretability across all evaluated architectures. For MobileNetV2, the best performance and the most compact attention maps were obtained with SE and ECA (reaching 91%), while Soft Attention improved accuracy but produced broader, less concentrated activation patterns. For ResNet50, SE achieved the most focused and symptom-specific heatmaps, whereas CBAM reached the highest classification accuracy (up to 90.4%) but generated more spatially diffuse Grad-CAM activations. Overall, these findings demonstrate that attention-enhanced CNNs can provide accurate, interpretable, and robust detection of palm tree diseases and pests under real-world agricultural conditions. Full article

Semantic change detection has become a key technology for monitoring the evolution of land cover and land use categories at the semantic level. However, existing methods often lack effective information interaction and fail to capture changes at multiple granularities using single-scale features, resulting in inconsistent outcomes and frequent missed or false detections. To address these challenges, we propose a three-branch model HCTANet, which enhances spatial and semantic feature representations at each time stage and models semantic correlations and differences between multi-temporal images through three innovative modules. First, the multi-scale change mapping association module extracts and fuses multi-resolution dual-temporal difference features in parallel, explicitly constraining semantic segmentation results with the change area output. Second, an adaptive collaborative semantic attention mechanism is introduced, modeling the semantic correlations of dual-temporal features via dynamic weight fusion and cross-temporal cross-attention. Third, the spatial semantic residual aggregation module aggregates global context and high-resolution shallow features through residual connections to restore pixel-level boundary details. HCTANet is evaluated on the SECOND, SenseEarth 2020 and AirFC datasets, and the results show that it outperforms existing methods in metrics such as mIoU and SeK, demonstrating its superior capability and effectiveness in accurately detecting semantic changes in complex remote sensing scenarios. Full article

Background/Objectives: Accurate segmentation of skin lesions is essential for early skin cancer detection. However, traditional CNNs are limited in modeling long-range dependencies, leading to poor performance on lesions with complex shapes. Methods: We propose MSDTCN-Net, a dual-encoder network that integrates ConvNeXt and Deformable Transformer to extract both local details and global semantic information. A Squeeze-and-Excitation (SE) mechanism is introduced to adaptively emphasize important channels. To address scale variation in lesions, we design a Multi-Scale Receptive Field (MSRF) module combining multi-branch and dilated convolutions. Furthermore, a Hierarchical Feature Transfer (HFT) mechanism is employed to guide high-level semantics progressively to shallow layers, enhancing boundary reconstruction in the decoder. Results: Extensive experiments on the ISIC 2016, ISIC 2017, ISIC 2018, and PH2 datasets show that MSDTCN-Net achieves competitive performance across metrics including IoU, Dice, and ACC, validating its effectiveness and generalization in skin lesion segmentation. Conclusions: MSDTCN-Net effectively combines local and global feature extraction, multi-scale adaptability, and semantic guidance to achieve high-accuracy skin lesion segmentation, demonstrating its potential in clinical diagnostic applications. Full article

Sustainable finance exhibits non-linear dynamics, regime shifts, and distributional drift that challenge conventional forecasting, particularly in volatile emerging markets. Conventional models, which often overlook this structural complexity, can struggle to produce stable or reliable probabilistic forecasts. To address this challenge, this study introduces a complexity-aware forecasting framework that operationalizes information-theoretic meta features, Shannon entropy (SE), permutation entropy (PE) and Kullback–Leibler (KL) divergence to make Environmental, Social, and Governance (ESG) index forecasting more stable, probabilistically accurate, and operationally reliable. Applied in an emerging-market setting using Türkiye’s ESG index as a natural stress test, the framework was benchmarked against a macro-technical baseline with a calibrated XGBoost classifier under a strictly chronological, leakage-controlled nested cross-validation protocol and evaluated on a strictly held-out test set. In development, the framework achieved statistically significant improvements in both stability and calibration, reducing fold-level dispersion (by 40.4–66.6%) across all metrics and enhancing probability-level alignment with Brier score reduced by 0.0140 and the ECE by 0.0287. Furthermore, a meta-analytic McNemar’s test confirmed a significant reduction in misclassifications across the development folds. On the strictly held-out test set, the framework’s superiority was confirmed by a statistically significant reduction in classification errors (exact McNemar p < 0.001), alongside strong gains in imbalance-robust metrics such as BAcc (0.618, +12.8%) and the MCC (0.288, +38.5%), achieving an F1-score of 0.719. Overall, the findings of the complexity-aware framework indicate that explicitly representing the market’s informational state and transitions yields more stable, well-calibrated, and operationally reliable forecasts in regime-shifting financial environments, supporting enhanced robustness and practical deployability. Full article

We propose MUSE++, an advanced and lightweight speech enhancement (SE) framework that builds upon the original MUSE architecture by introducing three key improvements: a Mamba-based state space model, dynamic SNR-driven data augmentation, and an augmented multi-objective loss function. First, we replace the original multi-path enhanced Taylor (MET) transformer block with the Mamba architecture, enabling substantial reductions in model complexity and parameter count while maintaining robust enhancement capability. Second, we adopt a dynamic training strategy that varies the signal-to-noise ratios (SNRs) across diverse speech samples, promoting improved generalization to real-world acoustic scenarios. Third, we expand the model’s loss framework with additional objective measures, allowing the model to be empirically tuned towards both perceptual and objective SE metrics. Comprehensive experiments conducted on the VoiceBank-DEMAND dataset demonstrate that MUSE++ delivers consistently superior performance across standard evaluation metrics, including PESQ, CSIG, CBAK, COVL, SSNR, and STOI, while reducing the number of model parameters by over 65% compared to the baseline. These results highlight MUSE++ as a highly efficient and effective solution for speech enhancement, particularly in resource-constrained and real-time deployment scenarios. Full article

Oceanic whitecaps are caused by wave breaking and are very important in air–sea interactions. Usually, whitecap coverage is considered a key factor in representing the role of whitecaps. However, the accurate identification of whitecap coverage in videos under dynamic marine conditions is a tough task. An EMA-SE-ResUNet deep learning model was proposed in this study to address this challenge. Based on a foundation of residual network (ResNet)-50 as the encoder and U-Net as the decoder, the model incorporated efficient multi-scale attention (EMA) module and squeeze-and-excitation network (SENet) module to improve its performance. By employing a dynamic weight allocation strategy and a channel attention mechanism, the model effectively strengthens the feature representation capability for whitecap edges while suppressing interference from wave textures and illumination noise. The model’s adaptability to complex sea surface scenarios was enhanced through the integration of data augmentation techniques and an optimized joint loss function. By applying the proposed model to a dataset collected by a shipborne camera system deployed during a comprehensive fishery resource survey in the northwest Pacific, the model results outperformed main segmentation algorithms, including U-Net, DeepLabv3+, HRNet, and PSPNet, in key metrics: whitecap intersection over union (IoU_W) = 73.32%, pixel absolute error (PAE) = 0.081%, and whitecap F1-score (F1_W) = 84.60. Compared to the traditional U-Net model, it achieved an absolute improvement of 2.1% in IoU_W while reducing computational load (GFLOPs) by 57.3% and achieving synergistic optimization of accuracy and real-time performance. This study can provide highly reliable technical support for studies on air–sea flux quantification and marine aerosol generation. Full article