Search Results (14,763)

Traffic accident detection is a critical component of intelligent transportation systems (ITS), enabling timely incident response and traffic management. While most existing approaches rely on temporal information from video sequences, such methods are not always applicable in resource-constrained surveillance environments. This study investigates the feasibility of detecting traffic accidents from single static images by formulating the task as a binary classification problem. Representative architectures, including Vision Transformer (ViT), Swin Transformer, and ResNet-50, are systematically evaluated on the Car Crash Dataset (CCD) under multiple training configurations. To assess generalization capability, cross-domain evaluation is conducted using an external crash video dataset (ECVD) constructed to approximate real-world deployment conditions. Experimental results show that all models achieve strong performance under in-domain evaluation. However, cross-domain testing reveals substantial performance degradation, particularly in recall, indicating limited generalization capability under domain shift. Qualitative analysis further shows that missed detections are associated with weak visual cues, occlusion, and complex traffic environments, while false positives are caused by visually ambiguous patterns resembling accident scenarios. Unlike prior studies that primarily report performance improvements, this work provides empirical evidence that model behavior in static-image-based accident detection is governed by dataset composition rather than architectural design. Therefore, static-image-based accident detection should be interpreted as a coarse-level screening tool rather than a fully reliable decision-making system. This study highlights the importance of data-centric design and cross-domain evaluation for improving real-world applicability. Full article

The dynamics of initial layer-by-layer growth and subsequent nucleation of quantum dots of Si and Ge on Si(001) were studied combining reflection high-energy electron diffraction, scanning electron microscopy and atomic force microscopy. It was shown that the processes occurring at the initial stage determine further growth of the heterostructure and final shape and density of nanoislands. The mechanisms of terrace formation, occurrence and dynamics of dimer rows of the 2 × N superstructure, and effects of temperature on the growth characteristics were described. The obtained experimental dependences show the critical relationship between the synthesis parameters (growth temperature), epitaxial growth processes and the characteristics of the resulting nanoislands. The fundamental studies conducted make it possible to create self-organizing quantum dots of a given size and density for advanced optoelectronics, including infrared photosensitive elements and single-photon detectors. Full article

Extreme disasters often induce multiple simultaneous faults in distribution networks, posing significant risks to power supply reliability. Although network reconfiguration and intentional islanding are critical strategies for enhancing system resilience, existing studies typically address them separately and fail to adequately account for the uncertainties associated with renewable energy generation and load demand. To address these limitations, this paper presents a collaborative optimization model for resilient distribution network restoration. A multi-time-step dynamic restoration framework is developed to coordinate network reconfiguration, emergency repair scheduling, distributed generation dispatch, and load shedding. This framework enables unified decision-making for island formation and topology reconfiguration, and incorporates an island integration mechanism to broaden the feasible solution space. To manage source–load uncertainties, chance-constrained programming is introduced, transforming probabilistic security constraints into deterministic equivalents using risk indicator variables, thereby striking a balance between operational security and economic efficiency. In addition, the model optimizes repair sequences under multi-fault conditions to enhance resource utilization. Simulations on a modified IEEE 33-node system validate the effectiveness of the proposed approach in reducing load curtailment, accelerating restoration, and achieving a favorable trade-off between operational risk and economic performance. Full article

Structural colors generated by interference, diffraction, or light scattering offer vivid visual effects without dyes or electronic components, making them promising for flexible optical sensing. This work reports a simple stretch–plasma–release (S-P-R) strategy to fabricate wrinkled photonic elastomers (WPEs). The flexible periodic structures exhibit mechanically responsive structural colors, as tensile strain alters the grating period, generating optical signals that can be visualized and quantified by spectroscopy. The wrinkle period is tunable in the range of 0.4–3.42 μm by adjusting plasma power, exposure time, pre-stretch ratio, and film thickness. A dumbbell-shaped substrate design reduces edge-induced stress concentration. It shows improved wrinkle uniformity, with the coefficient of variation reduced from 6.64% to 2.74%, and experimental colors agreeing well with modified Bragg condition predictions. The reflection peak shows a significant shift from 356 nm to 658 nm with varying viewing angles. Patterned plasma treatment enables the selective generation of wrinkled structures, producing bright color patterns. The structural color can be fully erased at a critical strain of 20% and recovered upon release, remaining stable over multiple loading–unloading cycles. With excellent mechanical compliance and optical tunability, these materials are well-suited for integration with hydrogel-based systems and show promise for wearable devices, security marking, and anti-counterfeiting applications. Full article

Pine wilt disease (PWD) is a devastating biological forest disturbance, making its large-scale and high-precision remote sensing monitoring crucial for epidemic prevention and control. However, the performance of existing deep learning methods in high-resolution imagery is often limited by the confusion of spectral features among disparate ground objects and the complexity of forest boundaries. To address these challenges, this study proposes an innovative, end-to-end deep learning architecture termed MBA-Former. Built upon the robust Swin Transformer V2 backbone, the model systematically integrates two highly adaptable functional modules: (1) a front-end intelligent fusion module designed to adaptively fuse heterogeneous features, and (2) a back-end boundary refinement module that refines segmentation contours via dual-task learning. To train and evaluate the model, fine-grained manual annotations were first performed on Gaofen-2 satellite imagery acquired from multiple typical epidemic areas across northern and southern China. Information-enhanced datasets were constructed by fusing the original spectral bands, typical vegetation indices, and texture features. A comprehensive performance evaluation was then conducted, specifically targeting typical challenging scenarios characterized by complex ground object boundaries. The experimental results demonstrate that the Multi-modal Boundary-Aware Transformer (MBA-Former) significantly outperforms current state-of-the-art models. It achieved a mean Intersection over Union (mIoU) of 81.74%, an IoU of 77.58% for the most critical infected tree category, and a Boundary F1-Score of 78.62%. Compared to the best-performing baseline model, Swin-Unet, these three metrics exhibited notable improvements of 2.88%, 3.55%, and 4.46%, respectively. These findings convincingly demonstrate that MBA-Former provides a highly accurate and robust solution for the large-scale, automated remote sensing monitoring of forest diseases, offering immense value in preventing significant economic losses and preserving forest ecosystem integrity. Full article

Wireless sensor networks increasingly support time-critical monitoring applications, where coverage optimization must often be performed under limited computational resources. This work addresses a previously underexplored WSN coverage problem involving range-free, angular-limited sensors with transmitter-induced sensing degradation and discrete sector orientation. We formulate a mixed combinatorial problem that jointly optimizes K-out-of-N sensor activation and sector assignment under strict feasibility constraints. A constraint-aware genetic algorithm with repair-based feasibility enforcement is proposed and validated against the global optimum obtained via exhaustive enumeration, enabling direct quantification of optimality. The repair mechanism corrects infeasible offspring after each genetic operation to guarantee that exactly K sensors remain active, eliminating the need for penalty-based constraint handling. A brute-force search is used to establish the global optimum of our small-scale scenario, serving as a ground-truth optimality benchmark for evaluating the proposed method. The purpose of this comparison is not to assess competitiveness against other metaheuristic algorithms, but to quantify how closely the proposed approach approximates the true optimal solution under strict problem constraints. The constraint-aware genetic algorithm is developed using an integer chromosome encoding, two initialization strategies, two crossover pairing schemes, elitism, and per-gene mutation, combined with alternative constraint-handling strategies. Two experimental series evaluate the impact of population size, crossover method, mutation probability, and constraint handling using problem-specific metrics, alongside convergence and fitness statistics. The proposed algorithm reliably reaches near-optimal solutions with significantly reduced computational cost when compared to exhaustive search. By integrating problem-specific constraints directly into the process, the proposed evolutionary optimization method effectively balances solution quality and execution time, making it well suited for scenarios requiring rapid sensor reconfiguration. Full article

Traditional wood identification models are vulnerable to out-of-distribution (OOD) substitution in the global timber trade. In response to this issue, this study presents a dual-stage cascade authentication architecture using in situ Raman spectroscopy and machine learning. First, a physically informed preprocessing strategy, integrating adaptive truncation (>1749 cm⁻¹) and first-derivative filtering, is developed to extract a 1309-dimensional composite feature matrix. This step effectively decouples non-linear fluorescence and converts physical detector saturation into highly discriminative features. To mitigate data leakage, the system utilizes a cross-validated Random Forest engine for Stage-1 closed-set discriminative screening. Subsequently, it cascades a high-dimensional One-Class Support Vector Machine (OCSVM) for Stage-2 open-set non-linear boundary verification in the Reproducing Kernel Hilbert Space. This design avoids the “variance trap” of traditional linear dimensionality reduction (e.g., PCA), preserving weak but critical secondary metabolite signals. Under a controlled OOD benchmarking scenario involving three taxonomically and chemically similar substitute species, the optimized Stage-1 engine maintains a 91.67% closed-set accuracy on known species. Crucially, Stage-2 verification achieves an open-set detection AUROC of 0.9722 and limits the FPR95 to 3.33%. Feature importance mapping indicates that the model effectively incorporates macroscopicoptical surrogate features (e.g., fluorescence decay boundaries) for decision-making. Overall, this study offers a robust, controlled non-destructive approach for real-world wood authenticity verification. Full article

Background/Objectives: Efficient management of intensive care unit (ICU) resources is a critical challenge for modern healthcare systems, which must balance high-quality patient care with operational and financial performance. ICU length of stay (LOS) is a key metric of clinical complexity and hospital efficiency. However, traditional methods for predicting LOS often fail to capture the complex, nonlinear interactions among physiological, demographic, and treatment-related variables. Machine learning (ML) and deep learning (DL) models have emerged as promising tools for enhancing predictive accuracy and supporting data-driven decision-making. Methods: This study presents a systematic review and meta-analysis of ML and DL approaches for predicting ICU LOS in adult patients. Following PRISMA guidelines, eight scientific databases were searched, yielding 33 eligible studies published between 2015 and 2025. Results: Mixed medical–surgical ICUs were the most common setting (51.5%), and 45.5% of datasets were sourced from public repositories. Most studies (19/33) focused on binary classification of prolonged stays, although thresholds ranged from >48 h to ≥14 days. The pooled results from ten studies yielded an AUROC of 0.9005 (95% CI: 0.8890–0.9121), indicating strong predictive capability across diverse clinical contexts. Subgroup analyses showed comparable performance between specialized surgical and general ICUs. Conclusions: These findings suggest that AI-driven LOS prediction models exhibit strong discriminatory power for ICU LOS prediction, supporting hospital capacity planning. However, to translate this into reliable clinical support, the methodological heterogeneity, scarcity of external validation, and near absence of calibration reporting identified in this review need to be addressed. Full article

Aphids are major agricultural pests worldwide, causing crop damage both through direct piercing-sucking feeding and the transmission of plant viruses. Their multistage life cycle, unique developmental physiology, plasticity in developing pesticide resistance, and multifaceted interactions with host plants and bacterial endosymbionts make effective control particularly challenging. In this review, we summarize the current toolbox available for aphid control across major crop systems, including chemical pesticides, biological agents, plant resistance, cultural practices, biorational control, and emerging strategies such as RNA interference (RNAi) and symbiosis-targeted approaches. Rather than providing an exhaustive survey of the literature, we draw on conceptual and illustrative studies to critically evaluate the strengths and limitations of each control strategy. Finally, we outline future directions for aphid control, highlighting the potential of modern technologies, such as artificial intelligence (AI), synthetic biology, data-driven analytics, and CRISPR-based genome editing, to expand and improve existing control options. Full article

Shale gas, as an unconventional resource, requires hydraulic fracturing to create complex fracture networks due to its low porosity and permeability. However, the presence of interlayers significantly affects fracture propagation, leading to highly complex fracture morphologies. This study focuses on the interbedded shale of the WJP Formation in southern China. A three-dimensional block discrete element method (BDEM) was employed to establish a hydraulic fracture propagation model, systematically investigating the effects of geological parameters (stress difference, interlayer thickness), engineering parameters pumping rate, fluid volume, viscosity), and perforation parameters (cluster number, cluster spacing, perforation location) on fracture network morphology. The results indicate that: (1) Among geological parameters, interlayer thickness is the key factor inhibiting vertical fracture propagation. Due to the influence of interlayers, an increase in stress difference promotes fracture length but suppresses fracture height and stimulated reservoir volume (SRV); (2) For engineering parameters, there exists a “threshold effect” for pumping rate and fluid volume, with 16 m³/min and 2000 m³ identified as the critical thresholds for interlayer breakthrough. Low viscosity (1 mPa·s) is conducive to forming complex fracture networks, while high viscosity extends fracture length but reduces SRV; (3) Regarding perforation parameters, the optimal stimulation effect is achieved with 6–7 clusters, a cluster spacing of 10 m, and perforation locations in the center of the main shale layer (19.85–21.6 m); (4) By introducing grey relational analysis, the degree of correlation between various influencing factors and the response to interlayer breakthrough is systematically evaluated based on the breakthrough conditions under different factors. Thin interlayers or low stress differences can reduce the critical pumping rate, whereas thick interlayers (≥3 m) become the primary constraint, making breakthrough difficult even at high pumping rates. Reliable interlayer breakthrough requires the simultaneous satisfaction of Δσ ≤ 16 MPa, h < 1 m, and Q ≥ 16 m³/min. The reliability of the model was verified by comparing numerical simulation results with field microseismic data. This study reveals the extension laws of complex fracture networks in interbedded shale, providing a theoretical basis for fracturing design and development optimization. Full article

Digestive tract tumors represent a predominant contributor to the global public health burden, with conventional therapeutic modalities experiencing inherent limitations and immunotherapy being impeded by the immunosuppressive property and heterogeneity of the tumor microenvironment (TME). This makes the gut microbiota–immune axis a promising therapeutic target. Marine polysaccharides, endowed with distinctive structural characteristics, exhibit potential in the modulation of this regulatory axis, yet their structure–activity relationships (SARs) and the intrinsic limitations in delivery efficiency remain largely unelucidated. In this review, we systematically synthesized the latest research advances pertaining to the modulation of the gut microbiota–immune axis by marine polysaccharides in digestive tract tumors, in accordance with the logical framework of polysaccharide structure, flora regulation, immune activation, tumor inhibition, and delivery optimization. We elaborated on the bidirectional crosstalk between the gut microbiota and the immune axis during tumorigenesis, as well as the regulatory effects and core underlying mechanisms of marine polysaccharides derived from algal, animal and microbial sources on this axis, including targeted floral regulation, microbiota-mediated immune activation, and direct/indirect tumor suppression. We also analyzed the key structural determinants and structural modification strategies of marine polysaccharides, alongside the development of nanodelivery systems for the improvement of their oral bioavailability. Furthermore, we identified critical existing research gaps, such as the ambiguous SARs and poor oral bioavailability of marine polysaccharides, and propose the integration of multi-omics analysis, synthetic biology technology and advanced nanodelivery strategies as the core future research directions in this field. Collectively, marine polysaccharides hold tremendous promise as novel therapeutic agents for digestive tract tumors, and interdisciplinary collaboration is regarded as indispensable for their successful clinical translation and translational application. Full article

The gut microbiota serves as a critical interface for host immunity, making it a promising target for probiotic intervention. In this study, we investigated the immunomodulatory potential of the strain Bifidobacterium breve (B. breve) MN15965 and the underlying role of gut bacterial communities in this process. We first assessed its in vitro immunomodulatory activity by measuring nitric oxide and cytokine secretion in THP-1 macrophages. Subsequently, an immunosuppressed mouse model was established by treating BALB/c mice with cyclophosphamide (CTX), a chemotherapeutic agent known to cause immune dysfunction and mucosal damage. In this model, we performed a series of analyses, including H&E staining, measurement of hematological parameters and serum cytokines/immunoglobulins, quantification of fecal short-chain fatty acids (SCFAs) by gas chromatography, and profiling of gut microbiota composition via 16S rRNA gene amplicon sequencing. The results showed that MN15965 supernatant enhanced TNF-α, IL-1β, and GM-CSF secretion in THP-1 cells, promoting M1 macrophage activation in vitro. In the in vivo model, MN15965 administration restored spleen and thymus tissue integrity and improved physiological indices, hematological parameters, and immunoglobulin levels. Furthermore, MN15965 increased fecal SCFAs, particularly butyric and valeric acid, increased gut bacterial diversity, and enriched potentially beneficial SCFA-producing taxa, including Lachnospiraceae and Eubacterium. These findings demonstrate that B. breve MN15965 alleviated CTX-induced immunosuppression by activating immune responses, regulating gut bacterial communities, and boosting SCFA production. Full article

Bees are essential pollinators for agricultural systems, making accurate, automated monitoring of their behavior critical for assessing colony health and ecosystem stability. Recent advances in computer vision and artificial intelligence have enabled large-scale bee traffic monitoring at hive entrances; however, most existing event classification methods focus exclusively on simple entrance and exit events. This simplification overlooks compound movements—such as U-turns and guarding behaviors—that represent a substantial portion of bee activity and can lead to inaccurate trajectory reconstruction and misleading behavioral interpretations. In this work, we systematically analyze existing event classification strategies used in automatic bee traffic monitoring, evaluating their performance on both simple and compound movements. We then propose extended classification methods that explicitly model compound events by incorporating bidirectional movement patterns derived from positional and angular cues. Using a manually annotated dataset of computer-vision-based hive entrance recordings, we compare threshold-based, displacement-based, and angle-based approaches under simple and mixed-event conditions. Our results demonstrate that compound events account for over one-third of all detected movements and that classification methods explicitly designed to handle bidirectional behavior substantially outperform traditional approaches in both accuracy and robustness. In particular, threshold-based bidirectional classification achieves near-perfect performance when full trajectories are available, while displacement-based methods provide a reliable alternative under partial observations. These findings highlight the importance of modeling compound behaviors in automated bee monitoring systems and contribute to more accurate flight reconstruction, behavioral analysis, and AI-driven decision support for precision agriculture and pollinator management. Full article

Edible insects have emerged as a sustainable source of high-quality proteins, lipids, and carbohydrates (including chitin), as well as micronutrients such as minerals and vitamins, and diverse bioactive compounds, thereby making them promising ingredients for functional food applications. Their favourable nutritional profile and low environmental footprint make them attractive ingredients for next-generation food systems. However, processing and preservation remain critical challenges, particularly with respect to the stability of bioactive compounds, lipid oxidation, and protein functional properties such as solubility, emulsifying capacity, and water-holding capacity. This review critically examines recent advances in food processing technologies applied to edible insects, including drying, extraction, fermentation, and microencapsulation, with emphasis on their effects on bioactive compound retention and functional performance. The role of processing strategies in enhancing oxidative stability, protein solubility, emulsifying properties, and overall technological applicability is discussed, alongside safety, regulatory, and consumer acceptance considerations. Overall, this review highlights key technological pathways for the effective valorisation of insect-derived ingredients and outlines future directions for their integration into sustainable and functional food products. In contrast to previous reviews, this work provides a comparative and mechanism-oriented analysis of processing methods, highlighting inconsistencies across studies and identifying key technological trade-offs. Particular attention is given to the relationship between processing parameters and the stability of bioactive compounds. Full article

Autonomous task assignment in multi-unmanned aerial vehicle (UAV) systems operating in dynamic and safety-critical airspace environments is highly challenging due to complex spatial interactions and rapidly changing relative geometries. This paper proposes a hierarchical decision-making framework that bridges individual maneuvering behaviors with cooperative task allocation in multi-agent aerial systems. First, a high-fidelity single-agent maneuver model is learned using a physics-consistent simulation environment, where spatial advantage is evaluated based on relative distance and angular relationships within a kinematically feasible interaction zone (KIZ). Subsequently, a Geometry-Aware Graph Attention Network (GA-GAT) is developed to address scalable multi-agent assignment problems. Unlike conventional approaches that rely on flat feature representations, the proposed method explicitly incorporates kinematic feasibility constraints into the attention mechanism via a novel gating module, enabling efficient relational reasoning under dynamic conditions. The proposed framework is applicable to a range of civilian and safety-oriented scenarios, including UAV swarm coordination, emergency response monitoring, infrastructure inspection, and autonomous airspace management. Simulation results demonstrate that the GA-GAT-based approach significantly outperforms heuristic baselines in terms of coordination efficiency and overall system performance in complex multi-agent environments. This study highlights that decoupling maneuver-level control from high-level coordination provides a scalable and computationally efficient solution for real-time multi-UAV decision-making in safety-critical applications. The proposed framework is designed for general multi-agent coordination problems in civilian aerial applications. Full article