Search Results (9,268)

Maritime transport contributes 2.9% of global greenhouse gas emissions, traditionally evaluated through operational fuel cycles while neglecting lifecycle impacts. This study redefines merchant ship recycling as a strategic front-end pillar for global decarbonization by assessing Embodied Carbon Trade-offs. Utilizing a mixed PRISMA systematic and semi-systematic methodology of literature from 2019–2025, we analyzed bulk carriers, container ships, and tankers across six thematic clusters. Our findings demonstrate that scenarios involving the potential decommissioning of up to 22.2% of the global merchant fleet exceeding 20 years of age could significantly mitigate lifecycle emissions by displacing primary iron-ore smelting with circular electric arc furnace marine-steel recovery. Crucially, this environmental dividend is non-linear and bound by regional energy matrices; under deeply decarbonized grids, this technological displacement can theoretically yield an upper-bound 72% emissions reduction, whereas fossil-heavy power supplies significantly diminish net mitigation margins. Practically, this research provides an operational roadmap for shipowners and regulators navigating the Hong Kong Convention and carbon border mechanisms. This work concludes that sustainable shipbreaking has the potential to function as an economically viable, strategic reservoir of low-carbon raw materials essential for achieving international net-zero targets throughout the entire shipping structural lifecycle. Full article

People counting is one of the key tasks in intelligent monitoring systems. However, accurately counting people in dynamic environments can be extremely challenging. This is especially true in mobile robot applications, where challenges such as a moving camera, varying conditions, and a limited field of view due to environmental obstacles arise. In such scenarios, the people counting task primarily involves visually detecting and grouping individuals to determine the total number of unique people. This paper presents a people counting algorithm based on visual people detection and clustering. The method utilizes the You Only Look Once (YOLO) detector to identify the bounding boxes of detected individuals and extract features from their corresponding regions of interest (ROIs). Additionally, the dimension of the extracted features is reduced using an encoder and clustered to distinguish individuals, with the number of clusters serving as an estimate of the number of people. The method was tested on a dataset containing 45 independent sequences with a total of 19,350 RGB images, complete with metadata for people detection and re-identification. This dataset encompasses various settings, particularly scenarios featuring mobile robots moving and capturing frames in indoor environments. The experiments demonstrate the effectiveness of the proposed method in different configurations. The best results were achieved using the YOLOv10n detector combined with K-means or SK-means clustering, yielding a Mean Absolute Error ($MAE$) of 1.11. The proposed encoder-based method significantly reduces clustering time and operates effectively within the limited resources available on mobile robotic platforms such as the Jetson Nano. Full article

The bean bug, Riptortus pedestris (Fabricius), and the southern green stink bug, Nezara viridula (L.), are important pod-sucking pests of soybean, Glycine max (L.) Merr. Their feeding on pods and developing seeds induces soybean staygreen symptoms and causes severe yield losses. To assess their future damage risk, we integrated MaxEnt, PCA env and a phenological matching index (PMI) using global occurrence records, bioclimatic variables and soybean phenology data. Under current climatic conditions, R. pedestris exhibited a predominantly temperate East Asian distribution, with suitable areas extending farther north and northeast. By contrast, N. viridula showed a more southerly and spatially continuous distribution across South and Southeast Asia, with suitability declining markedly toward northern and northeastern Asia. By the 2090s under SSP5-8.5, suitable areas decreased by 26.7% for N. viridula but increased by 34.0% for R. pedestris, and the co-suitable area declined from 9.460 to 7.350 million km². PCA env indicated low to moderate niche overlap (Schoener’s D = 0.278) with significant niche differentiation. Under fixed soybean calendars, mean PMI generally increased for both pests, although the increase plateaued for R. pedestris under high emission scenarios by the late 21st century. Collectively, these findings indicate that Asian soybean production regions currently face overlapping damage risk, that the potential damage zone of R. pedestris is likely to expand further, and that the temporal overlap between the soybean sensitive period and adult activity windows of both pests will generally increase—despite the projected contraction in the potential damage distribution of N. viridula. Full article

This study extends an established bi-level dynamic optimization framework for grid-connected regional integrated energy systems (RIES) to address the escalating renewable energy penetration (REP) within integrated power systems (IPS). While traditional models treat REP as static, our approach integrates its dynamic growth into both the design and operational scheduling phases, utilizing a genetic algorithm paired with the Gurobi solver. Applied to a case study in Changsha, the extended model is systematically benchmarked against conventional static REP scenarios. The research shows that the introduction of REP growth factors in the optimization model can increase the installation capacity of ground source heat pumps (GSHPs), reduce the installation capacity of combined heat and power units and absorption chillers, and reduce the initial investment of the system by 12.20%. Affected by the difference in equipment capacity configuration, the primary energy consumption and total cost of the grid-connected RIES decrease during the planning period, while the cumulative carbon dioxide emissions show a slight increase. The primary energy consumption and total cost under winter operating conditions were reduced by 2.44% and 2.38%, respectively. In addition, with the increase of REP growth rate in IPS, the reductions in the system’s primary energy consumption, carbon dioxide emissions, and total cost all increase accordingly. Therefore, in macroscopic RIES planning, incorporating dynamic REP reveals a clear trade-off: improving economic and energy efficiency may temporarily increase carbon emissions when renewable penetration is low. Full article

This study addresses the energy demand profiles of commercial buildings by developing an optimal dispatch strategy for a regional high-efficiency distributed energy system integrating electricity, cooling, and storage through source–load coordination. The spatiotemporal distribution characteristics of cooling, heating, and electrical loads are analyzed and an integrated energy system model is established, comprising gas internal combustion engine, a lithium bromide absorption chiller/heater, gas-fired boiler, centrifugal chillers, and an ice storage system. Taking into account seasonal electricity pricing policies and meteorological variations in Shanghai, a load grading system and a time-of-use (TOU) pricing response mechanism are constructed, leading to the development of operational strategy portfolios for different typical scenarios. A multi-objective optimization dispatch model is formulated with the dual aims of minimizing operating costs and maximizing energy efficiency. The results indicate that, compared to a fixed operational mode, the optimized strategy achieves average CO₂ emission reduction rates of 16.4% in summer, 25.2% in non-summer periods, and 20.9% annually. Additionally, annual grid electricity purchases are reduced by 10.2%, with a static investment payback period of 11.37 years. This research provides an intelligent, practically applicable operational solution for distributed energy systems in commercial buildings, effectively overcoming the limitations of traditional approaches in terms of flexibility and economic performance. Full article

Single-image dehazing aims to recover clear scenes from degraded images affected by atmospheric scattering, serving as a critical preprocessing technique for improving the imaging quality of visual sensors. Existing deep learning-based dehazing methods exhibit limited generalization ability in real-world scenarios, primarily due to the spatial non-uniformity of haze and its coupling with illumination and texture degradation, as well as the scarcity of real paired data. To address these issues, this paper proposes a haze-density conditional attention fusion network (DCAF-Net). The network employs an adaptive haze density perception module to fuse priors such as the dark channel, local contrast, and saturation, generating a spatial haze density guidance map. This map is then embedded as conditional information into the multi-scale feature modulation and attention fusion process, enabling adaptive restoration of regions with different degradation levels. Furthermore, a residual dense cascaded feature enhancement module is designed to leverage feature reuse, gated fusion, and residual learning to enhance the representational capacity of deep features. Training adopts a joint optimization objective combining Charbonnier reconstruction loss, perceptual contrast loss, and structural similarity loss. Experimental results demonstrate that DCAF-Net achieves competitive performance against representative methods on multiple synthetic and real-world hazy datasets, and shows promising restoration performance on representative real-world hazy scenes, and can provide high-quality image preprocessing support for visual-sensor-based intelligent perception systems. Full article

Ground-based radar monitoring of low Earth orbit (LEO) satellites is constrained by terrain occlusion and short visibility windows. Deploying UAVs as mobile platforms can extend coverage, but beam alignment, obstacle avoidance, and kinematic limits have not been jointly addressed. This paper proposes a UAV path planning framework that treats the ground station–UAV–satellite (G-U-S) collinear relationship as a planning prior. Ideal UAV positions are first computed from satellite visible arcs. In obstacle-free airspace, these positions directly define the flight path. When no-fly zones intersect the ideal trajectory, a geometry-guided RRT* planner generates collision-free detours by favoring sampling near the beam axis. The beam safety corridor is then embedded with kinematic limits as a conservative hard constraint within a Minimum Snap Bézier QP model, whose convex hull property bounds the trajectory within the coverage region. Simulations over a representative visible arc of the target satellite (approximately 3.3 min) achieve 100% beam coverage in the obstacle-free scenario and 95.2% with three no-fly zones placed along the ideal trajectory. Visible arc duration varies with satellite orbital trajectory; the framework is applicable to any arc whose geometry satisfies the kinematic constraints of the UAV platform. An ablation experiment confirms that the geometry-guided sampling strategy is essential: replacing it with uniform random sampling reduces coverage by 9.5 percentage points and triggers a diagnostic fallback mechanism. Edge case stress testing further identifies the conditions under which beam containment and obstacle avoidance come into tension. The proposed framework provides a simulation-validated trajectory planning approach for UAV-assisted LEO satellite monitoring in complex airspace. Full article

This study examines the optimal coordination of internal and external rescue for troubled banks under systemic contagion. Using annual data for 210 Chinese commercial banks from 2013 to 2024, it constructs a region-constrained minimum-density interbank network and embeds it in an EN-GLT dual-channel contagion framework that captures both direct default losses and asset fire-sale losses. Each bank is sequentially treated as the initially shocked institution, and pure internal rescue, pure external rescue, and mixed rescue strategies are compared under risk-tolerance, rescue-capacity, cost, and moral-hazard constraints. The results show that capital-loss contagion and fire-sale amplification are economically meaningful under the no-rescue scenario and become stronger as the fire-sale markdown rate rises. Mixed rescue outperforms pure internal or pure external rescue in most years, with the optimal internal rescue share mainly concentrated between 30% and 55%. The findings indicate that problem-bank resolution should combine internal loss absorption with external stabilization and should be differentiated according to contagion channels, bank type, and the nature of the crisis. Full article

Identifying the impact location (“sweet spot”) on a tennis racket is crucial for performance evaluation in tennis training. However, existing approaches typically rely on expensive vision-based systems or specialized sensors, limiting their applicability in real-world scenarios. We propose a sound-sensor-based multi-task framework for racket impact localization using acoustic signals, combining radial region classification with continuous position regression. To effectively model complex acoustic patterns, we design a multi-expert convolutional neural network (CNN) architecture with multi-scale feature extraction and task-specific optimization. Each expert branch operates at a different temporal receptive field and is trained with tailored loss functions, enabling complementary learning of global patterns, class imbalance characteristics, and hard samples. The shared backbone jointly supports both classification and regression tasks, allowing the model to learn more informative and structured representations. Experimental results demonstrate that the proposed framework consistently outperforms conventional methods in radial region classification while achieving accurate impact position estimation. Furthermore, additive noise augmentation significantly improves robustness, enabling stable performance under noisy and practical sensing conditions. Full article

Object detection in real-world scenarios is often challenged by adverse visual conditions, such as low illumination, strong glare, and dense fog, which severely degrade visible-spectrum features and lead to missed detections, inaccurate localization, and reduced detection accuracy. To address these issues, this paper proposes AQFS-Net, a dual-modal fusion detection network for visible-infrared object detection. Built upon YOLOv13, AQFS-Net adopts a symmetric dual-branch backbone by incorporating infrared images, thereby exploiting the complementary information between the visible and infrared modalities. To alleviate the negative transfer caused by conventional static fusion strategies, an Adaptive Quality-Aware Fusion Module (AQFM) is designed to dynamically enhance informative features and suppress degraded information according to the modality-specific reliability of different regions. In addition, a Foreground-Aware Saliency Guidance (FASG) branch is introduced to guide the network to focus on target regions through foreground supervision, reducing interference from complex backgrounds. Experimental results on the public LLVIP and M3FD datasets show that the proposed method improves mAP@0.5 by 6.8 and 3.1 percentage points, respectively, compared with the baseline using only visible images. These results demonstrate the effectiveness of AQFS-Net in improving dual-modal fusion quality and detection performance under challenging visual conditions, providing a practical reference for visible-infrared object detection in complex illumination scenarios. Full article

The super-resolution (SR) of Magnetic Resonance Imaging (MRI) is essential for utilizing clinical scans with limited resolution, noise, and anisotropic sampling, such as multi-modal brain tumor imaging. In this work, we propose a Tri-Representation hybrid framework for MRI SR, TRT-GLA, that redefines the MRI SR task as a joint spatial–spectral–structural high-resolution image generation problem. TRT-GLA utilizes (i) spatial global–local attentions for modeling the spatial anatomy, (ii) a Fourier spectral transfer mechanism for upholding spectral consistency, and (iii) multi-scale hierarchical spectral decomposition for improved edge details. To adapt the learning framework to medical imaging characteristics, we introduce a tri-representation consistent loss function that explicitly combines pixel-wise, spectral, and edge structure priors from the high-resolution ground-truth, as well as a progressive resolution learning strategy. Our large-scale brain tumor experiments, on the IXI, BraTS 2019, 2020, and 2023 datasets, show that TRT-GLA achieves state-of-the-art results at upsampling factors of ×2, ×4, and ×8, respectively, achieving substantial improvements across CNN, GAN, and transformer-based methods in Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Multi-scale Structural Similarity Index (MS-SSIM). We further demonstrate how SR benefits brain tumor segmentation through the downstream task evaluation of a dual-branch segmentation framework. TRT-GLA produces highly accurate tumor segmentation results from low-resolution inputs, improving over native high-resolution inputs at ×8 in critical tumor boundary regions and in small tumor regions. There remains a small gap between native, high-resolution imaging and SR-enhanced performance, which TRT-GLA nearly closes under realistic scenarios. Our results highlight the importance of synthesizing unified priors over spatial, spectral, and structural domains within a transformer for anatomically faithful reconstructions. Importantly, we also establish the utility of TRT-GLA in supporting quantitative analysis through a downstream tumor segmentation experiment that is clinically relevant. Full article

Catalyst deactivation shifts the optimal operating region of exothermic fixed-bed reactors, yet most reactor digital twins focus on monitoring rather than catalyst-state-aware operating decisions. This work presents a simulation-based self-optimizing digital-twin prototype integrating a physics-based reactor model, a moving-window constrained activity estimator, and a target-optimization layer for o-xylene oxidation to phthalic anhydride in a vanadia–titania heat-exchanged fixed-bed reactor. Sparse axial temperature and conversion measurements are reconciled to estimate an axial catalyst activity profile; gas and coolant inlet temperatures are then updated subject to a hot-spot safety constraint. The estimator achieved an activity-profile root mean square error (RMSE) of 0.075, an outlet-conversion RMSE of 0.99 percentage points, and an outlet-temperature RMSE of 1.85 K. Under the baseline noisy-measurement scenario, estimated activity optimization raised the mean phthalic anhydride yield from 46.3% under fixed targets to 61.9%, within 0.14 percentage points of the true-activity optimum, while maintaining the maximum reactor temperature below 730 K. In this matched-model simulation study, this corresponds to recovering approximately 99.1% of the yield improvement available with perfect catalyst-state knowledge. The policy remained superior to fixed-target operation across all tested noise levels, sensor configurations, and kinetic pre-exponential perturbations. All results are obtained from synthetic-measurement simulations rather than experimental or plant data, and plant validation is still required to quantify structural model error. The findings demonstrate the value of linking catalyst-state estimation to operating-target adaptation in a reproducible catalytic-reactor digital-twin workflow. Full article

Regional carbon balance emerges from the complex interplay between land-use dynamics, spatial economic activities, and ecological processes as a socio-ecological system cannot be captured by any single analytical lens. This study develops an integrated assessment workflow that links three components: (i) coefficient-based carbon emission and sequestration accounting by land-use type; (ii) intra-provincial spatial-interaction analysis operationalized through two complementary tools—the Ecological Support Coefficient (ESC) and Economic Contribution Coefficient (ECC), which characterize the local economy–ecology relationship within each city, and a gravity-based emission-linkage model that uses GDP, population, emissions, and inter-city distance to characterize the network structure of inter-city emission attraction; and (iii) the intPLUS model, which combines random-forest-derived transition probabilities with patch-generation rules to simulate multi-scenario land-use trajectories. The framework is applied to Jiangsu Province, China, across 13 prefecture-level cities, using 1995–2020 historical data and three 2030 scenarios. Model performance is validated against observed 2020 land use, with an overall Kappa coefficient of 0.82. The results reveal a stable “high-south–low-north” gradient in emissions, a contrasting “high-ECC/low-ESC” versus “low-ECC/high-ESC” combining pattern across southern and northern Jiangsu, and a hierarchical core–periphery emission-linkage network anchored on the southern metropolitan cluster. Scenario projections for 2030 show clear divergence in provincial carbon budgets, with emissions of $\mathrm{12,326.59}\times {10}^4$ t, $\mathrm{11,745.26}\times {10}^4$ t, and $\mathrm{13,243.42}\times {10}^4$ t under natural development, ecological protection, and economic development, respectively, and corresponding sequestration of $-87.10\times {10}^4$ t, $-100.29\times {10}^4$ t, and $-85.61\times {10}^4$ t. Rather than treating carbon accounting and land-use simulation in isolation, this workflow bridges the analytical gap between physical land-use transitions and socioeconomic spatial emission spillovers, translating structural interactions into actionable spatial planning strategies. Relying on widely available data, the framework demonstrates strong methodological transferability for comparable subnational systems, provided that local parameters and sink coefficients are properly recalibrated. Full article

Accurate carbon emission prediction is critical for regional low-carbon transitions and the realization of China’s “dual carbon” goals. However, carbon emission systems exhibit significant complex nonlinear relationships and time-dependent characteristics, making it difficult for traditional statistical methods and conventional neural networks to fully capture their dynamic evolution. To address this challenge, this paper proposes a multivariate carbon emission prediction method that integrates a Temporal Dendritic Neural Model (TDNM) with a Dendritic Adaptive Learning (DAL) algorithm. Based on panel data from 54 prefecture-level cities in China spanning 1999 to 2023, this study systematically constructs a carbon emission driving factor system comprising 13 indicators across economic, social, and energy dimensions. Empirical results indicate that the proposed model achieves competitive predictive performance and improved stability compared with several conventional machine learning models and deep learning baselines. Furthermore, taking the carbon emission predictions of Suzhou City as a starting point, a detailed scenario analysis is conducted to clarify the evolution trends of future carbon emissions under different development pathways, providing methodological references for analyzing carbon emission evolution and low-carbon development pathways in industrial cities. Full article

The COVID-19 pandemic accelerated the use of telemedicine and revealed persistent barriers in rural primary healthcare, especially among older adults and caregivers. This study proposes an LLM-mediated smart tele-primary-care framework to support respiratory infection monitoring in underserved communities of Córdoba, Colombia. The framework is based on caregiver-centered teleconsultation, remote monitoring, preventive education, structured symptom reporting, risk-based teletriage, and clinician-supervised digital support. The LLM functions as a controlled conversational interface to collect symptoms, organize patient information, reinforce health education, generate follow-up reminders, and identify predefined warning signs, without replacing clinical judgment or making autonomous diagnostic or treatment decisions. A preliminary scenario-based assessment examined whether the proposed workflow supports coherent triage and appropriate escalation of high-risk cases. Importantly, the described architecture—a telematic system for intelligent-engine access in assisted medicine support (ATIMAMS)—has been implemented and is currently operational at the Systems Engineering Program, Universidad del Sinú (Montería, Colombia), providing intelligent decision support for telemedicine consultations and remote assistance appointments in the study region. Overall, the study presents a context-sensitive, low-barrier, and safety-aware model for strengthening rural primary care, improving continuity of care, and supporting caregiver-mediated respiratory infection monitoring. Full article