Search Results (25,275)

Existing self-supervised methods have achieved significant success in ground-level autonomous driving scenarios, but applying them directly to Unmanned Aerial Vehicle (UAV) videos remains challenging. On the one hand, rapid pose changes in UAVs often lead to oblique-view imaging, making it difficult for conventional methods to handle the perspective distortion in oblique imagery. On the other hand, complex UAV viewpoints may cause depth blurring in low-texture regions. To address these challenges, we propose a lightweight self-supervised monocular depth estimation method for UAV scenarios. By utilizing a Dynamic Direction-Aware Module (DDaM), the network adaptively adjusts the sampling grid to correct distorted features during feature extraction, while enhancing its ability to capture features at different spatial locations. Furthermore, to mitigate the loss of spatial information caused by multiple downsampling operations, we integrate a Coordinate Attention Mechanism into the encoder. This mechanism captures features along two separate spatial axes, preserving the spatial coordinates of object boundaries. Our experiments demonstrate that the synergy between DDaM and the Coordinate Attention Mechanism enables the prediction of more accurate object boundaries and richer local details. To validate the effectiveness and practical applicability of the proposed method, we conduct experiments on both the MidAir synthetic dataset and the UAVid real-world dataset. The results show that, compared with current baseline methods, our approach maintains competitive performance while requiring the fewest parameters. Full article

Cracks are among the most common pavement distresses, and their timely detection is crucial for road maintenance. Existing methods struggle to completely capture elongated and irregular cracks, often resulting in fragmented detection outputs, which leads to the inaccurate assessment of crack length and affects the reliability of pavement condition evaluation. To address this issue, this paper proposes a dual-path crack segmentation network that integrates CNN and Transformers. The CNN branch incorporates a dynamic multi-branch convolution module to enhance the directional perception and structural modeling of elongated cracks. The Transformer branch employs a lightweight DCNv4 module to replace traditional self-attention mechanisms, effectively capturing long-range dependencies while reducing computational complexity. A multi-path fusion module is designed to achieve the collaborative enhancement of dual-path features, improving the semantic representation of continuous crack regions. Additionally, a combined loss function of BCE and Dice is adopted to alleviate the severe class imbalance between crack and background pixels, further improving the completeness of crack segmentation. Experiments on four datasets, including CFD, DeepCrack537, Gaps384, and Crack500, demonstrate that the proposed model outperforms all compared methods in terms of F-score and mIoU. Ablation studies further validate the effectiveness of the dual-path architecture and its key modules in improving performance. Furthermore, in field validation on real road scenarios, the pavement condition index (PCI) calculated based on the proposed method shows an average deviation of only 0.81 compared to manually interpreted ground truth, demonstrating the practical value of continuous crack detection for pavement maintenance assessment. Full article

Basalt fiber-reinforced polymer (BFRP) composites are increasingly considered as a sustainable alternative to traditional FRP systems for the strengthening of reinforced concrete (RC) structures, owing to their favorable mechanical properties, durability, and lower environmental impact. This study investigates the effectiveness of externally bonded BFRP strips for the strengthening of RC beam–column joints, with particular attention to the influence of strengthening layout on the structural response. An experimental program was carried out on full-scale RC beam–column joint specimens subjected to monotonic loading with load–unload cycles of increasing amplitude. Each specimen was first tested in its original configuration to induce controlled damage and subsequently strengthened using BFRP strips arranged according to two different layouts. This approach enabled a direct comparison between the behaviour of pre-damaged and retrofitted specimens and allowed the contribution of the BFRP reinforcement to be clearly identified. BFRP strengthening markedly improves joint performance, enhancing strength, ductility, and energy dissipation while limiting stiffness degradation. The results underline the critical role of the strengthening layout in governing the effectiveness of the composite system, as well as the influence of substrate cracking in the activation of the BFRP reinforcement. Full article

With recent advances in deep reinforcement learning (DRL) and graph neural networks (GNNs), graph-based autonomous exploration methods have significantly improved decision-making performance in complex environments. However, under partial observability and sparse-reward conditions, existing methods still struggle with long-horizon decision-making and sustained exploration. To address these challenges, we propose a spatiotemporal graph learning framework, termed ST-GICM, that improves the robustness and efficiency of autonomous exploration by integrating graph-structured encoding, temporal memory, and an intrinsic curiosity mechanism. Specifically, a Graph Attention Network (GAT) and a Spatiotemporal Reasoning Core (STRC) are employed to dynamically encode the viewpoint graph and fuse temporal memory, thereby alleviating perceptual aliasing in graph-based exploration. In addition, an Intrinsic Prediction Module (IPM) is designed to generate intrinsic rewards based on the prediction error of graph-level latent representations, thereby encouraging sustained exploration. Experiments conducted in procedurally generated complex topological environments show that the proposed method outperforms existing baselines in terms of coverage rate, success rate, repeated revisit rate, and oscillation count, while maintaining trajectory costs comparable to those of the baselines. These results demonstrate the effectiveness and superiority of ST-GICM in partially observable environments under sparse-reward conditions. Full article

Essential oils have attracted increasing attention due to their bioactive properties. This review focuses on their anticonvulsant potential by exploring the relation between the chemical composition of essential oils and the mechanism of action underlying this effect. Evidence from in vivo and ex vivo studies is presented to identify structure–activity relations and to distinguish well-supported effects from preliminary findings. Moreover, as essential oil’s quality is vital to ensure safety and efficacy in pharmacotherapeutic approaches. For this reason, factors including extraction and analytical methods as well as authenticity assessment are discussed due to their impact on pharmacological consistency and reproducibility. Overall, this review highlights key compounds and mechanisms contributing to anticonvulsant activity, identifies current limitations in the literature, and outlines priorities for future research aimed at validating essential oils as potential complementary therapeutic agents in seizure management. Full article

Extracellular DNA (eDNA) has gained attention as a danger signal between organisms because of the ecological implications of this mechanism and its great potential as a biological modulator in agriculture. Self-DNA and non-self DNA have been evaluated earlier, both as plant immune system elicitors. Here we show the effect of eDNA extracted from the bacterial phytopathogen Clavibacter michiganensis applied to tomato plants in different concentrations (50, 100 and 150 µg mL⁻¹). Monitoring morphology of the plants, spectrophotometric determinations and RT-qPCR assays showed a dose-dependent effect on plant growth and root development, activation of antioxidant enzymes such as catalase and superoxide dismutase, biosynthesis of secondary metabolites, including phenolic compounds and flavonoids, and differential expression of genes related to plant stress response, such as chalcone synthase and phenylalanine ammonia-lyase. Lower concentration treatments showed an increment in the variables as beneficial responses for agricultural practices, and the higher concentration (150 µg mL⁻¹) showed reduced or no effects on the evaluated variables. This work represents a step forward in the development of effective and more sustainable agricultural technology in crop production. Full article

Acetylcholine (ACh) is an evolutionarily conserved neurotransmitter that is widely distributed in the central and peripheral nervous systems and plays essential roles in multiple physiological processes. This review summarizes the full biological cycle of ACh, including its synthesis, vesicular storage, release, degradation, and reuptake, and discusses the regulatory mechanisms underlying its functions in the nervous system and peripheral organs. Through nicotinic acetylcholine receptors (nAChRs) and muscarinic acetylcholine receptors (mAChRs), ACh is involved in central nervous system functions such as cognition, learning and memory, attention, arousal, reward, and decision-making, as well as peripheral processes including motor control, autonomic regulation, and immune modulation. In addition, ACh plays a pivotal role in the brain–body axis. At the central level, the nervous system regulates peripheral organ function through autonomic and neuroendocrine pathways. At the peripheral level, cholinergic signals derived from the enteric nervous system and immune cells convey information about the body’s internal state to the central nervous system through vagal and other afferent pathways, forming an important bottom-up regulatory network. Collectively, these findings indicate that ACh is not only a classical neurotransmitter but also a key molecular mediator of brain–body communication. A more comprehensive understanding of cholinergic signaling may provide new insights into physiological regulation and the pathogenesis of neurological, psychiatric, cardiovascular, and inflammatory diseases. Full article

Network traffic classification is a cornerstone of network management and security. Addressing the challenges of feature extraction in encrypted traffic and the deployment limitations of traditional deep learning models on resource-constrained edge devices due to their large parameter sizes, this paper proposes a lightweight network for encrypted traffic classification, termed CEMA-Net (Convolutional Positional Encoding and Efficient Multi-scale Attention Network). Specifically, the proposed model integrates an Efficient Multi-scale Attention (EMA) mechanism with a Convolutional Positional Encoding (CPE) strategy to jointly capture global dependencies and local contextual information. To enable efficient adaptation to traffic data, an Efficient Multi-scale Attention Adapter (EMAAdapter) is designed, which reconstructs one-dimensional traffic sequences into a pseudo-2D representation and extracts horizontal, vertical, and local features in parallel. This design facilitates effective modeling of complex cross-scale dependencies in encrypted traffic with minimal computational overhead. Experimental results on three public datasets demonstrate that the proposed method, with only 0.66 M parameters, achieves superior classification performance compared with mainstream vision-based models such as ResNet-101, while significantly reducing computational cost. These results highlight the effectiveness of combining convolutional positional encoding with multi-scale attention mechanisms and provide an efficient solution for encrypted traffic classification in resource-constrained environments. Full article

Battery Energy Storage Systems (BESSs) are increasingly deployed in grid-scale applications, electric mobility, and renewable integration, where safety, reliability, and longevity are critical. Thermal runaway remains one of the most severe failure modes in lithium-ion batteries, often triggered by complex interactions between electrochemical, thermal, and mechanical phenomena. This paper presents an extended hybrid Physics-Informed Neural Network (PINN) framework for thermal anomaly prediction and early detection of runaway precursors in BESS. The proposed architecture integrates governing physical laws, specifically the Bernardi heat generation equation and Fick’s diffusion law, within a deep learning pipeline composed of a physics module, a temporal Bi-LSTM, and an attention mechanism for explainability, which may represent an obstacle in the application of deep learning algorithms. Beyond the initial formulation, the extended version presented here provides a deeper theoretical background, an expanded methodological justification, a more comprehensive comparison with state-of-the-art approaches, and a detailed discussion on scalability, uncertainty, and deployment challenges. The results for synthetic yet physically consistent datasets represent a proof of concept of the PINN approach, which can achieve superior generalization, robustness to noise, and interpretability compared to purely data-driven baselines, achieving an accuracy above 90% and an AUC of 0.95. The framework contributes to proactive safety management in cyber-physical energy systems and establishes a foundation for real-time, physics-aware anomaly detection in safety-critical BESS applications, e.g., marine transportation contexts and port environments. Full article

Image generation of plant disease in the natural environment has always been a challenging task. Traditional methods applied in the image generation of plant disease are without sufficient diversity and detailed lesions. Thus, this paper applies an image harmonization method to generate diverse combinations of disease images by integrating different backgrounds and target regions to enhance diversity. To construct the dataset, we captured real disease images of soybean and rice in natural environments. Next, the Squeeze-and-Excitation (SE) attention mechanism was integrated into the domain verification network (DoveNet), together with a mask guide generator, to focus more attention on lesions. Two discriminators worked together to capture global and local features, ensuring the preservation of critical contextual information. Finally, the improved DoveNet achieved a MSE of 43.77, a PSNR of 33.02, and an SSIM of 0.9806, showing a reduction of 3.61 in the MSE, an increase of 0.50 in the PSNR, and a 2.49% improvement in the SSIM compared with the original DoveNet. Meanwhile, through visual Turing tests we confirmed that images generated using the improved DoveNet were of much better quality and more convincing. Full article

This research investigates the formation and behavior of sustainable crumb rubber-modified gypsum–cement–pozzolan (GCP) composites, with a view to their use in a broad concept for construction. GCP binders are gaining attention as a low-carbon replacement for Portland cement, and the addition of recycled rubber helps the achievement of circular economy goals and potentially increases durability. The present research evaluates the impact of crumb rubber (CR) on the mechanical strength, water absorption, dimensional stability, and freeze–thaw resistance of 3D-printed GCP-rubber composites. Composite blends of variable proportions of crumb rubber were prepared at constant binder ratios. Mechanical properties were defined by prism specimens (40 × 40 × 160 mm) by the flexural and compressive strengths, and deformation was determined by micrometers to measure longitudinal strain as a function of curing. Water absorption was determined prior to freeze–thaw cycling to define pore saturation. Durability was investigated using two approaches: (1) controlled freeze–thaw experiments on cube specimens, with XF1 grade performance achieved, and (2) ultrasonic pulse velocity (UPV) testing of specimens 3D-printed for assessing internal structural change after long-term frost exposure. Results showed that compressive strength decreased moderately (10–20%) with increasing rubber content from 17% up to 50%, while flexural strength improved up to 15%, showing the elastomeric action of CR. Water absorption was reduced by 5–8% in the rubber-modified blends due to the hydrophobic character of rubber. Deformation tests also confirmed minimum length variation (<0.02%) during curing. Freeze–thaw durability was enormously improved, and test specimens retained more than 95% of initial strength. UPV measurements detected only a relatively modest velocity drop (~50 m/s) after 36 days cycling with subsequent stabilization up to 200 days, demonstrating long-term internal structure with minimal progressive damage. In summary, the findings demonstrate that GCP composites with crumb rubber incorporated are printable, dimensionally stable, and capable of freeze–thaw degradation resistance. Despite a moderate loss of compressive strength, the balance of introduced durability and sustainability suggests their competence as viable materials for additive manufacturing in construction. Full article

Soil salinization is increasingly threatening global agricultural productivity and food security, currently affecting over 6% of the world’s land and one-third of irrigated areas. Tomato (Solanum lycopersicum L.), a major vegetable crop worldwide, exhibits moderate sensitivity to salinity, which limits both its yield and fruit quality. In recent years, epigenetic regulation has gained attention as a key mechanism enabling flexible and reversible control of gene expression without altering DNA sequences. This review synthesizes current knowledge on the epigenetic control of salt stress responses in tomato, focusing on three interconnected levels: DNA methylation dynamics, RNA-directed DNA methylation (RdDM), and histone modifications. We explore how DNA methyltransferases reshape the methylome under salinity, using examples such as PKE1 and SlGI to illustrate functional gene-body methylation. The RdDM pathway is discussed with emphasis on the unexpected role of SlAGO4A as a negative modulator of stress tolerance and the growing evidence for RdDM-mediated regulation of transcription factors. We also examine the balanced regulation of histone acetylation and deacetylation, highlighting the conserved role of GCN5 in maintaining cell wall integrity and the diverse functions of histone deacetylases, such as SlHDA1, SlHDA3, and SlHDA5, in stress adaptation. Additionally, insights from wild tomato species and grafting-induced epigenetic changes are presented, revealing new dimensions of stress memory. Collectively, these epigenetic mechanisms constitute a complex regulatory framework that integrates stress responses with growth and development, providing potential targets for epigenetic breeding of salt-tolerant tomatoes. Full article

CRC remains a major cause of cancer-related morbidity and mortality worldwide. In recent years, the gut microbiota has gained increasing attention in CRC research. Intestinal microbes are not passive bystanders in tumor development. They may promote persistent inflammation, disrupt epithelial barrier integrity, alter microbial metabolites, and affect host immune and signaling pathways. Emerging evidence also suggests that microbiota-related metabolites and microbial functional alterations may influence host epigenetic regulation, including DNA methylation and chromatin-associated signaling, thereby further shaping colorectal carcinogenesis. Together, these changes can create a microenvironment that favors tumor initiation and progression. Several bacterial species, including Fusobacterium nucleatum, Parvimonas micra, and Peptostreptococcus anaerobius, have been repeatedly associated with CRC. In contrast, beneficial commensal microbes and their metabolites, especially short-chain fatty acids, may help maintain intestinal homeostasis and limit tumor-promoting processes. Because the gut microbiota is strongly shaped by diet, lifestyle, and environmental exposure, regional differences are also relevant. This is particularly important in Sichuan, China, where distinctive dietary habits and environmental features may influence microbial patterns associated with CRC risk and disease behavior. This review summarizes the main mechanisms linking the gut microbiota to CRC, examines the regional context of Sichuan, China, and discusses current and emerging clinical strategies. These include dietary intervention, probiotics, fecal microbiota transplantation, and microbiome-informed approaches to prevention, diagnosis, and treatment. Full article

This study investigates the effect of mass-based replacement of natural coarse aggregate with electric arc furnace (EAF) slag on the performance of ordinary Portland cement (OPC) concrete. Replacement levels of 0%, 30%, 50%, and 100% were examined, with particular attention to the volumetric changes induced by the higher density of EAF slag, which leads to an increase in paste volume. Fresh, mechanical, durability-related, and microstructural properties were evaluated. Results show a continuous reduction in workability with increasing slag content, despite the increase in paste volume, indicating the dominant influence of aggregate morphology on rheological behavior. Mechanical performance exhibited a non-linear response. Within the tested series, the 50% replacement mixture showed the highest mean compressive and splitting tensile strengths; however, the compressive strength difference relative to the control mixture remained small and within typical experimental scatter. In contrast, water absorption decreased progressively, reflecting improved matrix densification. However, this densification did not translate into enhanced mechanical performance, highlighting a decoupling between durability-related indicators and strength. A screening-level CO₂ assessment further showed that reductions in aggregate-related emissions were offset by increased cement content associated with mass-based replacement. The results emphasize the importance of considering volumetric effects when interpreting the behavior and sustainability of slag-based concrete. Note: all strength comparisons are based on mean values from three-specimen sets without formal statistical testing and should be regarded as exploratory observations. Full article

This study presents a comprehensive review of more than 100 research papers on sign language recognition (SLR) published between 2020 and 2026. The analysis focuses on deep learning approaches applied to video-based SLR, including spatiotemporal feature extraction, temporal modeling, attention mechanisms, motion-based representations, hybrid frameworks, transfer learning methods and other methods. Particular attention is given to how these methods model spatiotemporal dynamics and capture subtle gesture characteristics in sign language communication. The review highlights several recent developments, such as the introduction of specialized datasets, the emergence of real-time recognition systems, and the integration of multimodal fusion strategies. At the same time, persistent challenges remain, including data scarcity in low-resource sign languages, limited linguistic standardization of datasets, and insufficient model interpretability. The findings underline the importance of developing scalable and generalizable models capable of handling diverse datasets and user variability. The distinct contributions of this review are fourfold: (1) a comprehensive synthesis of over 100 studies published between 2020 and 2026, covering the full spectrum of deep learning architectures for video-based SLR; (2) a structured six-category taxonomy enabling systematic cross-architectural comparison; (3) a comprehensive focus on low-resource sign languages, which remain underrepresented in the existing literature; and (4) a critical analysis of the current benchmark landscape for low-resource sign languages, identifying key gaps and outlining strategic directions for future dataset development. These contributions are intended to guide further research toward more robust, inclusive, and universally applicable SLR systems. Full article