Search Results (2,083)

Climate hazards increasingly disrupt schooling, revealing the limits of preparedness models that treat teachers only as implementers. This study reframes teacher empowerment as a climate-resilience capability and examines how governance arrangements enable (or constrain) hazard-ready education systems. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR), searches of Web of Science, Scopus, and Google Scholar (2000–2025) identified 53 eligible studies. Across diverse hazards and settings, the evidence converges on a governance-to-capability pathway: empowerment becomes resilient performance only when the delegated decision space is matched with financed capacity (time, training, contingency resources), timely risk information and functional communication/digital infrastructure, institutionalized cross-sector coordination (education–DRR–health–protection–local government), and learning-oriented accountability (after-action review and adaptive revision rather than punitive compliance). Reported outcomes include higher preparedness quality, earlier protective action, improved learning continuity and safeguarding, and more sustainable teacher well-being/retention. Predictable failure modes include mandate–resource mismatch, accountability overload, unstable centralization–autonomy dynamics, and inequitable empowerment distribution affecting rural schools, women, and contract teachers, and disability inclusion. The evidence gaps remain pronounced for chronic hazards (especially heat and wildfire smoke), high-vulnerability contexts (fragile/conflict settings and informal settlements), and standardized measures of equity, burden distribution, governance performance, and cost-effectiveness. Policies should prioritize integrated governance packages with explicit protection and equity safeguards. Full article

Emotion classification is a fundamental task in affective computing, with applications in human–computer interaction, mental health monitoring, and social media analysis. Although most existing methods formulate it as a flat classification problem, emotional expressions are inherently structured and grounded in semantic roles such as the emotion cue, stimulus, experiencer, and target. However, the relative contribution of these roles to emotion inference has not been systematically examined. Unlike prior models, we propose RolEmo, a role-aware framework for emotion classification that explicitly incorporates semantic role information. The framework employs a controlled role-masking strategy to analyze the contribution of individual roles, augments textual representations with external commonsense knowledge to capture implicit affective context, and applies supervised contrastive learning to structure the embedding space by bringing emotionally similar instances closer while separating opposing ones. We evaluate RolEmo on three benchmark datasets annotated with semantic roles. Experimental results demonstrate that RolEmo outperforms the strongest baseline across three datasets by up to $16.4\%$, $25.8\%$, and $23.2\%$ in the Full Text, Only Role, and Without Role settings, respectively. The analysis further indicates that the cue and stimulus roles provide the most reliable signals for emotion classification, with their removal causing performance drops of up to $6.2\%$ in macro f1-score, while experiencer and target roles exhibit more variable effects. These findings highlight the importance of structured semantic modeling and commonsense reasoning for robust and interpretable emotion understanding. Full article

The modeling and control of photovoltaic and other engineering systems highly depend on the accuracy of parameter identification. However, parameter extraction for photovoltaic equivalent models typically presents a high-dimensional, strongly nonlinear, and multimodal global optimization problem. Traditional analytical or gradient-based methods are sensitive to initial values and easily fall into local optima. To address this issue, this paper proposes a multi-strategy improvement teaching–learning-based optimization algorithm (SNTLBO). A social learning network structure with symmetric interaction topology is introduced into the classical TLBO framework to characterize the knowledge propagation relationships among individuals. Through this symmetric and balanced information exchange mechanism, learners can be guided not only by the teacher but also by multiple neighbors within the network, enabling more diverse and symmetric exploration of the search space and enhancing population diversity and global search capability. Furthermore, a teacher reputation mechanism is constructed, where historical performance is used to weight teacher influence, strengthening the guidance of high-quality solutions and accelerating convergence. Meanwhile, an adaptive teaching factor is designed to dynamically adjust the teaching intensity based on the distance between the teacher and students in the solution space, maintaining a dynamic balance (symmetry) between exploration and exploitation. To evaluate the performance of the proposed algorithm, SNTLBO is systematically compared with 11 advanced optimization algorithms on two benchmark test suites, CEC2017 (30D, 50D) and CEC2022 (10D, 20D). Non-parametric statistical tests are conducted to assess significance. The results demonstrate that SNTLBO shows competitive advantages in terms of convergence speed, solution accuracy, and stability. Finally, SNTLBO is applied to the parameter estimation of single-diode, double-diode, triple-diode, quadruple-diode, and photovoltaic module models. Experimental results show that the proposed algorithm achieves higher identification accuracy and robustness in terms of RMSE, IAE, and I–V/P–V curve fitting, verifying its effectiveness and practical value for complex global optimization and practical engineering applications. Full article

Existing general image inpainting works require the user to customize a mask to indicate the region to be inpainted. However, the mask is often hard to calibrate accurately in real-world applications, e.g., graffiti removal. Blind image inpainting aims to automatically restore the degraded image into the visually reasonable one without a priori mask to indicate the area to be repaired. So far, most proposed blind inpainting methods convert the task into general inpainting by predicting the mask before inpainting. However, these methods are highly dependent on mask prediction results, which may produce inferior inpainting results if the prediction is inaccurate. To address this issue, we propose a two-stage blind inpainting framework with two novel designs: (1) cross-space contrastive learning, to remove the noise in the degraded images and realize the automatic inpainting in the latent space by reducing the distance of the degraded images and the corresponding complete images in the latent space; and (2) mask-aware adversarial training, to minimize the mutual information between the inpainted feature and the noise. Extensive experiments prove that our blind inpainting framework performs better on multiple datasets than the state-of-the-art methods. Full article

While school bullying has received substantial academic attention, the specific roles of teachers as (co-)perpetrators or bystanders in (cyber-)bullying dynamics remain markedly underexplored—particularly in the Austrian context. This article foregrounds pupils’ perception of teachers’ involvement in (cyber-)bullying. Drawing on feminist perspectives and insights from digital and gender(-queer) geographies, as well as interdisciplinary (cyber-)bullying research, it explores how pupils perceive teachers’ involvement in bullying dynamics and how they believe it shapes the perceived severity, trajectories, and outcomes of (cyber-)bullying. In doing so, the article contributes a specific but underexplored perspective on power and violence in schools. The analysis is based on 41 written narratives produced by young people attending upper secondary vocational colleges in Austria. The findings reveal that pupils subjectively perceive teachers as taking on various roles in (cyber-)bullying dynamics, including preventers, (silent) accomplices, defenders, outsiders, and (co-)perpetrators. In these accounts, teacher involvement in bullying reinforces power hierarchies, intensifies victimisation, and intersects with peer bullying dynamics, creating a complex system of interrelated influences. The study highlights the intersectional nature of discrimination and bullying, showing how pupils’ identities are entangled with their embodied experiences of both teacher- and peer-perpetrated bullying. These findings suggest an urgent need for spatially and structurally informed reforms in school policies and teacher training programmes to address teacher-perpetrated bullying, raise awareness of teachers’ responsibility in peer bullying dynamics, and foster safer, more inclusive learning spaces for pupils in Austria. Full article

Climate change is a critical global challenge, and the building sector accounts for nearly 30% of global greenhouse gas (GHG) emissions, remaining a key target for mitigation. Indoor environments contribute significantly to GHG emissions, primarily through heating, cooling, lighting, and occupant-driven energy use. Indoor mapping, serving as the foundation for Digital Twins (DTs), provides a spatiotemporal framework that integrates sensor data with Building Information Modelling (BIM), Geographic Information Systems (GIS), and Internet of Things (IoT) to support energy-efficient, low-carbon building operations. This review examined the role of indoor mapping in understanding, modelling, and reducing GHG emissions in buildings. It synthesized current advancements in indoor spatial data acquisition, ranging from Light Detection And Ranging (LiDAR) and Simultaneous Localization and Mapping (SLAM) to deep learning-based floor plan extraction, and evaluated their contribution to improved indoor environmental analysis. The review highlighted emerging techniques, challenges, and gaps, particularly the limited integration of physical indoor spaces with virtual layers representing assets, occupants, and equipment. Addressing this gap requires embedding spatial modelling as an intermediate analytical layer that structures and contextualizes sensor data to support spatiotemporal decision-making. Overall, this review demonstrated that indoor mapping plays a critical role in transforming spatial information into actionable insights, enabling more accurate energy modelling, enhanced real-time building management, and stronger data-driven strategies for GHG mitigation in the built environment. Full article

The proliferation of Resident Space Objects (RSOs), including satellites, rocket bodies, and debris, poses escalating challenges for Space Situational Awareness (SSA). Optical light curves capture temporal brightness variations influenced by factors such as attitude variation, viewing geometry, and surface properties. When appropriately processed and analyzed, these data can support RSO characterization and classification. This paper presents a scoping review of machine learning (ML) and deep learning (DL) methods for RSO classification using light-curve data. From 297 peer-reviewed studies published between 2014 and 2025, a screened subset of 29 works is selected for detailed methodological comparison. We trace the methodological evolution from handcrafted feature engineering toward convolutional, recurrent, and self-supervised models that learn representations directly from photometric time series. An analysis of three publicly accessible databases, Mini Mega TORTORA, Space Debris Light-Curve Database, and Ukrainian Database, reveals pronounced class imbalance, with payloads comprising over 80% of observations. While models trained on simulated data routinely achieve 95 to 99% accuracy, performance on measured light curves degrades to 75 to 92%, exposing a persistent gap between simulation and observation. We further identify data scarcity, repeated observations of the same objects, and inconsistent evaluation protocols as key barriers to reproducible benchmarking. Future progress will require benchmark-ready, sensor-aware datasets spanning diverse orbital regimes and viewing geometries, alongside physics-informed and transfer-learning approaches that improve robustness across sensors and between synthetic and observational domains. Full article

In low-Earth-orbit (LEO) satellite networks, the need for intelligent parameter-adjustment strategies has become increasingly critical due to the presence of highly dynamic channel conditions, limited spectrum resources, and complex interference environments. In this paper, a method for optimizing LEO satellite communication links based on deep reinforcement learning (DRL) is proposed. Through the optimization of the transmit power, the modulation and coding scheme (MCS), the beamforming parameters, and the retransmission mechanisms, adaptive link control is achieved in dynamic operational scenarios. A multidimensional state space is constructed, within which the channel state information, the interference environment, and the historical performance metrics are integrated. The spatio-temporal characteristics of the channel are extracted by means of a hybrid neural architecture that incorporates a convolutional neural network (CNN) and a long short-term memory (LSTM) network. To effectively accommodate both continuous and discrete action spaces, a hybrid DRL framework that combines proximal policy optimization (PPO) with a deep Q-network (DQN) is employed, thereby enabling cross-layer optimization of the physical-layer and link-layer parameters. The results demonstrate that substantial improvements in throughput, bit error rate (BER), and transmit-power efficiency are achieved under severely time-varying channel conditions, which provides a new idea for resource management and dynamic-environment adaptation in satellite communication systems. Full article

Rapid and accurate recognition of radar active jamming is a prerequisite for cognitive electronic countermeasures. However, under complex electromagnetic environments with scarce training samples, existing deep learning models are prone to capturing spurious correlations induced by environmental confounders, resulting in notable performance degradation. To address this causal confounding issue, we propose a physics-guided variational causal intervention network (PG-VCIN). First, we reconstruct a structured causal model of jamming signal generation, decoupling observations into robust physical statistical features and sensitive time–frequency image representations. Physical priors are then leveraged to perform dynamic precision-weighted modulation of visual feature extraction, enforcing physical consistency at the representation learning stage. Second, we formulate deconfounding within an active inference framework and introduce a variational information bottleneck to optimize mutual information, thereby filtering out high-complexity redundant information attributable to confounders while preserving the essential causal semantics. Finally, we numerically approximate the causal effect by imposing dual intervention constraints in the latent space, including intra-class invariance and confounder invariance. Experiments on a semi-physical simulation dataset demonstrate that the proposed method achieves substantially higher recognition accuracy than several representative few-shot baselines in extremely low-sample regimes, validating the effectiveness of integrating physical mechanisms with causal inference. Full article

Accurate urban green space (UGS) mapping is essential for assessing urban ecosystem health and supporting sustainable development planning. However, deep learning-based UGS segmentation from Red–Green–Blue (RGB) remote sensing imagery faces two major challenges. First, the absence of near-infrared (NIR) information in RGB imagery hinders the ability to discriminate spectrally similar classes, such as vegetation and non-vegetation. Second, conventional convolutions with fixed receptive fields struggle to model the complex and irregular boundaries characteristic of UGS. To address these challenges, this study combined the Normalized Green–Red Difference Index with the Deformable Convolutional Network Lab (NGRDI-DCNLab) model, a semantic segmentation model tailored specifically for RGB-only imagery. Based on the DeepLabV3+ framework, the model introduced three core improvements: (1) The Normalized Green–Red Difference Index (NGRDI) was incorporated to compensate for the absence of NIR information, enhancing the spectral separability of vegetation pixels. (2) Standard convolutions in the decoder were replaced with deformable convolutions, enabling the network to more effectively adapt to irregular boundaries of UGS. (3) An NGRDI-weighted loss function was designed to assign higher weights to challenging samples and uncertain boundary regions, guiding the model toward more accurate edge delineation. Comprehensive evaluations on two public high-resolution datasets—the Wuhan Dense Labeling Dataset (WHDLD) and the Beijing subset of the Urban Green Space-1m dataset (UGS-1m_Beijing)—demonstrated that the NGRDI-DCNLab model outperformed existing popular deep learning models (like Unet++, etc.). Specifically, the deformable convolution effectively enhances the feature modeling capability for irregular boundaries; incorporating the NGRDI vegetation index as a fourth channel strengthens spectral feature representation and improves the distinction between vegetation and non-vegetation; and adding the dynamic NGRDI-weighted loss enables targeted learning for challenging samples. Through the synergistic effect of these three modules, the model achieves mean Intersection over Union (MIoU) scores of 84.77% and 77.66%, as well as F1-scores of 91.75% and 87.27%, on the WHDLD and UGS-1m_Beijing datasets, respectively. Furthermore, the model exhibited certain generalization capability on the unmanned aerial vehicle (UAV) dataset, the Urban Drone Dataset 6 (UDD6), attaining an MIoU of 87.43%. Our results confirm that high-precision UGS extraction is achievable using only RGB remote sensing imagery, providing a cost-effective and practical technical solution for refined urban governance and ecological monitoring. Full article

Deaf and Hard-of-Hearing (DHH) students face persistent barriers in geometry education due to instructional approaches that inadequately support visual communication and embodied learning. This study examined DHH students’ experiences with GeoMETriA, a metaverse-based geometry learning platform integrating sign language instruction, three-dimensional visualization, and avatar-mediated interaction. Guided by the Cognitive Affective Model of Immersive Learning (CAMIL), a multi-phase qualitative design was employed, including pre-workshop interviews with four special education teachers and post-workshop focus group discussions with seven DHH secondary students following a four-session learning workshop. The findings indicate that gamified activities and peer collaboration enhanced interest and sustained engagement, while avatar customization supported embodiment and a sense of presence. Students described progression from initial uncertainty to greater confidence through practice and scaffolded support. However, cognitive and usability challenges emerged, particularly concerning sign language video pacing, navigation complexity, and limited instructional scaffolding. The study contributes theoretically by extending CAMIL-informed interpretations to sign-supported metaverse learning, empirically by documenting how engagement, embodiment, and self-efficacy develop during immersive geometry learning, and practically by offering design implications including adjustable sign language delivery, structured scaffolding, and culturally responsive avatar options. These findings suggest that metaverse-based platforms hold promise for supporting DHH learners when accessibility and learner-centered principles are embedded as foundational design considerations. Full article

This study investigates the cooling effects of coastal wetland systems in Hue City, Vietnam. The analysis focuses on their riparian buffer zones, defined here as areas within 600 m of the wetland boundary. Landsat 8 imagery was used to derive land surface temperature (LST) from 1 March to 31 July 2025—a recent period marked by multiple heatwaves across the region. To assess the cooling performance of wetlands, data samples were collected within the buffer zones. A Light Gradient Boosting Machine was trained to characterize the relationship between cooling intensity and a set of influencing factors (e.g., distance to wetland boundary, land use/land cover, built-up density, and green space density). The model explains approximately 91% of the variation in cooling intensity around wetlands. Notably, a machine-learning-based simulation framework was proposed to attain insights into the cooling characteristics of the riparian zone. The result indicates a mean cooling effect of about 2 °C and an effective cooling distance of 210 m from the wetland boundary. Partial dependence analysis further reveals that increasing built-up density substantially weakens cooling performance and implies that, for the conditions observed in Hue City, maintaining built-up density near wetlands below roughly 45% is favorable for sustaining effective cooling of the blue space, as indicated by the model-based partial dependence analysis. Overall, the research findings provide a data-driven basis for informing urban planning and wetland management in Hue City to mitigate heat stress. Full article

With accelerating urbanization and a global emphasis on quality of life, the effects of the built environment on individual physiological and psychological well-being have become a critical research focus. However, existing studies remain fragmented in terms of theoretical perspectives, spatial scales, and methodological approaches, and a comprehensive synthesis of the physiological and psychological effects of the built environment is still lacking. This review adopts an interdisciplinary approach, integrating architecture, urban planning, landscape architecture, geography, and psychology to systematically review the literature on the health impacts of the built environment. Its findings indicate that the scope of the built environment has expanded from natural settings to residential areas, streets, and public spaces. Research scales have progressed from macro-level districts to streets and public spaces and further to micro-level physical environments. The impacts have extended from emotional responses to broader health and well-being outcomes, with increasing attention being given to specific population groups. Technological advances have shifted research paradigms from traditional surveys to approaches incorporating big data, machine learning, virtual reality, and physiological monitoring, enabling more precise analyses of links between spatial perception and emotional responses. This review identifies gaps in interdisciplinary integration, long-term monitoring, and the consideration of individual differences, highlighting the need for future studies to integrate multimodal data with theory-informed practice to support more human-centered, health-promoting built environments. Full article

Low Earth Orbit (LEO) satellite signals of opportunity (SOP) provide a viable positioning alternative in GNSS (Global Navigation Satellite System)-denied environments, yet their accuracy is fundamentally constrained by the low-quality orbital data typically available, such as SGP4 (Simplified General Perturbations model 4) predictions derived from Two-Line Elements (TLEs). To address this limitation, this paper proposes a dynamics-informed non-causal deep learning framework that enhances low-quality orbital data into high-fidelity trajectories for accurate SOP positioning. The proposed Non-Causal Dynamics-Informed Representation Temporal Convolutional Network (Non-Causal DIR-TCN) integrates phase space reconstruction and a Temporal Convolutional Network to explicitly model the chaotic dynamics inherent in LEO orbits, while relaxing the causality constraints of standard temporal convolutions to utilize both past and future context from the available SGP4 stream. Experimental results demonstrate that the framework significantly reduces orbit estimation errors and accelerates model convergence. When applied to LEO-SOP positioning, it achieves approximately 20% improvement in 2D positioning accuracy compared to conventional SGP4-based methods. This work effectively bridges the gap between accessible low-precision orbital data and high-accuracy state estimation, advancing the practical deployment of opportunistic signals for resilient positioning in challenging environments. Full article

Real-time path planning for Unmanned Surface Vehicles (USVs) in complex marine environments remains challenging due to unstructured environments, ocean current disturbances, and dynamic obstacles. This paper proposes an improved Hybrid Safety and Reward-Sensitive Twin Delayed Deep Deterministic Policy Gradient (H_RS_TD3) algorithm and constructs a high-fidelity simulation environment based on GEBCO bathymetric data and CMEMS ocean current data. The path planning problem is formulated as a Markov Decision Process (MDP), where the state space incorporates multi-beam radar perception, ocean current disturbances, and relative goal information, while the action space outputs continuous thrust and rudder commands subject to vehicle dynamics constraints. The proposed framework integrates a risk-aware hybrid safety decision architecture, a Trajectory Predictor Network (TPN), a Curvature-driven Advantage-based Prioritized Experience Replay (CDA-PER) mechanism, and an uncertainty-aware conservative Q-learning strategy to enhance navigation safety, sample efficiency, and policy stability. Comprehensive simulations demonstrate that, compared with baseline deep reinforcement learning methods, the proposed approach achieves faster convergence, improved stability, and competitive path efficiency while consistently maintaining sufficient obstacle clearance and millisecond-level inference latency, validating its effectiveness and practical feasibility for safe USV navigation in realistic dynamic marine environments. Full article