Search Results (1,300)

This study introduces Itô-RMSProp, a novel extension of the RMSProp optimizer inspired by Itô stochastic calculus, which integrates adaptive Gaussian noise into the update rule to enhance exploration and mitigate overfitting during training. We embed this optimizer within Gated Recurrent Unit (GRU) networks for stock price forecasting, leveraging the GRU’s strength in modeling long-range temporal dependencies under nonstationary and noisy conditions. Extensive experiments on real-world financial datasets, including a detailed sensitivity analysis over a wide range of noise scaling parameters ($\epsilon$), reveal that Itô-RMSProp-GRU consistently achieves superior convergence stability and predictive accuracy compared to classical RMSProp. Notably, the optimizer demonstrates remarkable robustness across all tested configurations, maintaining stable performance even under volatile market dynamics. These findings suggest that the synergy between stochastic differential equation frameworks and gated architectures provides a powerful paradigm for financial time series modeling. The paper also presents theoretical justifications and implementation details to facilitate reproducibility and future extensions. Full article

Investigating the optimal planting strategies for brassica vegetables under variable climatic conditions is essential for developing sustainable production systems in northern agricultural regions. However, comprehensive knowledge about how planting timing modulates growth, physiological responses, and yield parameters across different cultivars remains limited. We investigated vegetative development, root morphology, physiological efficiency, and marketable yield in six cauliflower cultivars (‘Amazing’, ‘Cheddar’, ‘Clementine’, ‘Flame Star’, ‘Snow Crown’, and ‘Vitaverde’) subjected to four planting dates (May 1, May 15, June 1, and June 15) across two growing seasons (2023–2024), followed by detailed morphological and physiological profiling. Planting date, cultivar selection, and seasonal variation significantly influenced all measured parameters (p < 0.001), with notable interaction effects observed for fresh root weight, stomatal conductance, water use efficiency, and yield components. Early planted cultivars consistently demonstrated superior performance under variable environmental conditions, maintaining higher growth rates, enhanced root development, and improved physiological efficiency, particularly ‘Flame Star’, ‘Snow Crown’, and ‘Cheddar’, compared to late-planted treatments. Recovery of optimal plant development was most pronounced at May planting dates, with early-established crops showing better maintenance of vegetative growth patterns and enhanced yield potential, including higher curd weights (585.7 g for ‘Flame Star’) and superior marketable grades. Morphological profiling revealed distinct clustering patterns, with early-planted cultivars forming separate groups characterized by elevated root biomass, enhanced physiological parameters, and superior yield characteristics. In contrast, late-planted crops showed reduced performance, indicative of environmental stress responses. We conclude that strategic early planting significantly enhances cauliflower production resilience through comprehensive optimization of growth, physiological, and yield parameters, particularly under May establishment conditions. The differential performance responses between planting dates provide insights for timing-based management strategies, while the quantitative morphological and physiological profiles offer valuable parameters for assessing crop adaptation and commercial viability potential under variable climatic scenarios in northern agricultural systems. Full article

Lightweight Single-Image Super-Resolution (SISR) faces the core challenge of balancing computational efficiency with reconstruction quality, particularly in preserving both high-frequency details and global structures under constrained resources. To address this, we propose the Periodically Enhanced Cascade Network (PECNet). Our main contributions are as follows: 1. Its core component, a novel Multi-scale Adaptive Feature Aggregation (MAFA) module, which employs three functionally complementary branches that work synergistically: one dedicated to extracting local high-frequency details, another to efficiently modeling long-range dependencies and a third to capturing structured contextual information within windows. 2. To seamlessly integrate these branches and enable cross-window information interaction, we introduce the Periodic Boundary Padding Shift (PBPS) mechanism. This mechanism serves as a symmetric preprocessing step that achieves implicit window shifting without introducing any additional computational overhead. Extensive benchmarking shows PECNet achieves better reconstruction quality without a complexity increase. Taking the representative shift-window-based lightweight model, NGswin, as an example, for ×4 SR on the Manga109 dataset, PECNet achieves an average PSNR 0.25 dB higher, while its computational cost (in FLOPs) constitutes merely 40% of NGswin’s. Full article

Crack segmentation in images plays a pivotal role in the monitoring of structural surfaces, serving as a fundamental technique for assessing structural integrity. However, existing methods that rely solely on RGB images exhibit high sensitivity to light conditions, which significantly restricts their adaptability in complex environmental scenarios. To address this, we propose a structure-aware progressive multi-modal fusion network (SPMFNet) for RGB-thermal (RGB-T) crack segmentation. The main idea is to integrate complementary information from RGB and thermal images and incorporate structural priors (edge information) to achieve accurate segmentation. Here, to better fuse multi-layer features from different modalities, a progressive multi-modal fusion strategy is designed. In the shallow encoder layers, two gate control attention (GCA) modules are introduced to dynamically regulate the fusion process through a gating mechanism, allowing the network to adaptively integrate modality-specific structural details based on the input. In the deeper layers, two attention feature fusion (AFF) modules are employed to enhance semantic consistency by leveraging both local and global attention, thereby facilitating the effective interaction and complementarity of high-level multi-modal features. In addition, edge prior information is introduced to encourage the predicted crack regions to preserve structural integrity, which is constrained by a joint loss of edge-guided loss, multi-scale focal loss, and adaptive fusion loss. Experimental results on publicly available RGB-T crack detection datasets demonstrate that the proposed method outperforms both classical and advanced approaches, verifying the effectiveness of the progressive fusion strategy and the utilization of the structural prior. Full article

Drought stress during the reproductive phase substantially reduces seed yield and quality, posing a major challenge to sustainable crop production under climate change. This study investigated the effects of drought stress at the flowering stage on selected biochemical and physiological parameters in 18 carrot accessions. To describe the long-term consequences of drought comprehensively, we examined seed quality parameters. Our analyses revealed that stress responses are highly dependent on the genotype and the parameter examined. Regarding antioxidant responses and potential tissue damage caused by drought, ‘Dolanka’, DC97, DC265, DC359, DC522, DC701, DC704, and DC720 exhibited the highest tolerance. The photosynthetic apparatus and pigments were maintained under stress in DC233, DC522, DC717, and DC728. Germination parameters served as reliable indicators of stress tolerance in DC97, DC359, DC432, DC522, DC701, and DC722 accessions. Based on these findings and detailed discussion of the results, we conclude that tolerance/sensitivity assessment of carrot genotypes should consider the holistic response of the plant rather than individual parameters. Through overall assessment, we recommended DC522 accession as the most drought-tolerant, given its enhanced ROS (Reactive Oxygen Species) scavenging mechanisms, increased chloroplast pigments accumulation, and superior germination parameters under drought conditions. Conversely, DC295 should not be cultivated under water-deficient conditions due to its impaired ability to detoxify ROS, altered photosynthetic activity, and disrupted seed germination under such conditions. These results collectively highlight the potential for selecting drought-tolerant carrot genotypes in breeding programs targeting improved seed performance under water-limited conditions, thereby supporting the development of resilient cultivars adapted to future climate challenges. Full article

Hydrogel microneedles (HMNs) act as non-invasive devices that can effortlessly merge with the human body for drug delivery and diagnostic purposes. Nonetheless, their improvement is limited by intricate and repetitive issues related to material composition, structural geometry, manufacturing accuracy, and performance enhancement. At present, there are only a limited number of studies accessible since artificial intelligence and machine learning (AI/ML) for HMN are just starting to emerge and are in the initial phase. Data is distributed across separate research efforts, spanning different fields. This review aims to tackle the disjointed and narrowly concentrated aspects of current research on AI/ML applications in HMN technologies by offering a cohesive, comprehensive synthesis of interdisciplinary insights, categorized into five thematic areas: (1) material and microneedle design, (2) diagnostics and therapy, (3) drug delivery, (4) drug development, and (5) health and agricultural sensing. For each domain, we detail typical AI methods, integration approaches, proven advantages, and ongoing difficulties. We suggest a systematic five-stage developmental pathway covering material discovery, structural design, manufacturing, biomedical performance, and advanced AI integration, intended to expedite the transition of HMNs from research ideas to clinically and commercially practical systems. The findings of this review indicate that AI/ML can significantly enhance HMN development by addressing design and fabrication constraints via predictive modeling, adaptive control, and process optimization. By synchronizing these abilities with clinical and commercial translation requirements, AI/ML can act as key facilitators in converting HMNs from research ideas into scalable, practical biomedical solutions. Full article

Detection of changes in remote sensing relies on clean multi-temporal images, but cloud cover may considerably degrade image quality. Cloud removal, a critical image-restoration task, demands effective modeling of long-range spatial dependencies to reconstruct information under cloud occlusions. While Transformer-based models excel at handling such spatial modeling, their quadratic computational complexity limits practical application. The recently proposed Mamba, a state space model, offers a computationally efficient alternative for long-range modeling, but its inherent 1D sequential processing is ill-suited to capturing complex 2D spatial contexts in images. To bridge this gap, we propose the multi-pattern scanning Mamba (MPSM) block. Our MPSM block adapts the Mamba architecture for vision tasks by introducing a set of diverse scanning patterns that traverse features along horizontal, vertical, and diagonal paths. This multi-directional approach ensures that each feature aggregates comprehensive contextual information from the entire spatial domain. Furthermore, we introduce a dynamic path-aware (DPA) mechanism to adaptively recalibrate feature contributions from different scanning paths, enhancing the model’s focus on position-sensitive information. To effectively capture both global structures and local details, our MPSM blocks are embedded within a U-Net architecture enhanced with multi-scale supervision. Extensive experiments on the RICE1, RICE2, and T-CLOUD datasets demonstrate that our method achieves state-of-the-art performance while maintaining favorable computational efficiency. Full article

Flexible upconversion (UC) devices, owing to their unique combination of high–efficiency optical energy conversion and mechanical flexibility, have attracted increasing attention in the fields of optoelectronics, wearable devices, flexible displays, and biomedical applications. However, significant challenges remain in balancing optical performance, mechanical adaptability, long–term stability, and scalable fabrication, which limit their practical deployment. This review systematically introduces five representative upconversion mechanisms—excited–state absorption (ESA), energy transfer upconversion (ETU), energy migration upconversion (EMU), triplet–triplet annihilation upconversion (TTA–UC), and photon avalanche (PA)—highlighting their energy conversion principles, performance characteristics, and applicable scenarios. The article further delves into the flexible transition of upconversion devices, detailing not only the evolution of the luminescent layer from bulk crystals and nanoparticles to polymer composites and hybrid systems, but also the optimization of electrodes from rigid metal films to metal grids, carbon–based materials, and stretchable polymers. These developments significantly enhance the stability and reliability of flexible upconversion devices under bending, stretching, and complex mechanical deformation. Finally, emerging research directions are outlined, including multi–mechanism synergistic design, precise nanostructure engineering, interface optimization, and the construction of high–performance composite systems, emphasizing the broad potential of flexible UC devices in flexible displays, wearable health monitoring, solar energy harvesting, flexible optical communications, and biomedical photonic applications. This work provides critical insights for the design and application of high–performance flexible optoelectronic devices. Full article

Unmanned Aerial Vehicle (UAV)-based aerial object detection has been widely applied in various fields, including logistics, public security, disaster response, and smart agriculture. However, numerous small objects in UAV aerial images are often overwhelmed by large-scale complex backgrounds, making their appearance difficult to distinguish and thereby prone to being missed by detectors. To tackle these issues, we propose a novel Contextual-Semantic Interactive Perception Network (CSIPN) for small object detection in UAV aerial scenarios, which enhances detection performance through scene interaction modeling, dynamic context modeling, and dynamic feature fusion. The core components of the CSIPN include the Scene Interaction Modeling Module (SIMM), the Dynamic Context Modeling Module (DCMM), and the Semantic-Context Dynamic Fusion Module (SCDFM). Specifically, the SIMM introduces a lightweight self-attention mechanism to generate a global scene semantic embedding vector, which then interacts with shallow spatial descriptors to explicitly depict the latent relationships between small objects and complex background, thereby selectively activating key spatial responses. The DCMM employs two dynamically adjustable receptive-field branches to adaptively model contextual cues and effectively supplement the contextual information required for detecting various small objects. The SCDFM utilizes a dual-weighting strategy to dynamically fuse deep semantic information with shallow contextual details, highlighting features relevant to small object detection while suppressing irrelevant background. Our method achieves mAPs of 37.2%, 93.4%, 50.8%, and 48.3% on the TinyPerson dataset, the WAID dataset, the VisDrone-DET dataset, and our self-built WildDrone dataset, respectively, while using only 25.3M parameters, surpassing existing state-of-the-art detectors and demonstrating its superiority and robustness. Full article

Background: Current fMRI (functional magnetic resonance imaging)-driven brain information decoding for visual image reconstruction techniques faces issues such as poor structural fidelity, inadequate model generalization, and unnatural visual image reconstruction in complex scenarios. Methods: To address these challenges, this study proposes a Dynamic Confidence Bayesian Adaptive Network (DCBAN). In this network model, deep nested Singular Value Decomposition is introduced to embed low-rank constraints into the deep learning model layers for fine-grained feature extraction, thus improving structural fidelity. The proposed Bayesian Adaptive Fractional Ridge Regression module, based on singular value space, dynamically adjusts the regularization parameters, significantly enhancing the decoder’s generalization ability under complex stimulus conditions. The constructed Dynamic Confidence Adaptive Diffusion Model module incorporates a confidence network and time decay strategy, dynamically adjusting the semantic injection strength during the generation phase, further enhancing the details and naturalness of the generated images. Results: The proposed DCBAN method is applied to the NSD, outperforming state-of-the-art methods by 8.41%, 0.6%, and 4.8% in PixCorr (0.361), Incep (96.0%), and CLIP (97.8%), respectively, achieving the current best performance in both structural and semantic fMRI visual image reconstruction. Conclusions: The DCBAN proposed in this thesis offers a novel solution for reconstructing visual images from fMRI signals, significantly enhancing the robustness and generative quality of the reconstructed images. Full article

To address the challenges posed by complex background interference, varying target sizes, and high species diversity in bird detection tasks in the Dongting Lake region, this paper proposes an enhanced bird detection model named Birds-YOLO, based on the YOLOv11 framework. First, the EMA mechanism is introduced to replace the original C2PSA module. This mechanism synchronously captures global dependencies in the channel dimension and local detailed features in the spatial dimension, thereby enhancing the model’s robustness in cluttered environments. Second, the model incorporates an improved RepNCSPELAN4-ECO module, by reasonably integrating depthwise separable convolution modules and combining them with an adaptive channel compression mechanism, to strengthen feature extraction and multi-scale feature fusion, effectively enhances the detection capability for bird targets at different scales. Finally, the neck component of the network is redesigned using lightweight GSConv convolution, which integrates the principles of grouped and spatial convolutions. This design preserves the feature modeling capacity of standard convolution while incorporating the computational efficiency of depthwise separable convolution, thereby reducing model complexity without sacrificing accuracy. Experimental results show that, compared to the baseline YOLOv11n, Birds-YOLO achieves a 5.0% improvement in recall and a 3.5% increase in mAP@0.5 on the CUB200-2011 dataset. On the in-house DTH-Birds dataset, it gains 3.7% in precision, 3.7% in recall, and 2.6% in mAP@0.5, demonstrating consistent performance enhancement across both public and private benchmarks. The model’s generalization ability and robustness are further validated through extensive ablation studies and comparative experiments, indicating its strong potential for practical deployment in bird detection tasks in complex natural environments such as Dongting Lake. Full article

Software-defined networking (SDN) has reformed the traditional approach to managing and configuring networks by isolating the data plane from control plane. This isolation helps enable centralized control over network resources, enhanced programmability, and the ability to dynamically apply and enforce security and traffic policies. The shift in architecture offers numerous advantages such as increased flexibility, scalability, and improved network management but also introduces new and notable security challenges such as Distributed Denial-of-Service (DDoS) attacks. Such attacks focus on affecting the target with malicious traffic and even short-lived DDoS incidents can drastically impact the entire network’s stability, performance and availability. This comprehensive review paper provides a detailed investigation of SDN principles, the nature of DDoS threats in such environments and the strategies used to detect/mitigate these attacks. It provides novelty by offering an in-depth categorization of state-of-the-art detection techniques, utilizing machine learning, deep learning, and federated learning in domain-specific and general-purpose SDN scenarios. Each method is analyzed for its effectiveness. The paper further evaluates the strengths and weaknesses of these techniques, highlighting their applicability in different SDN contexts. In addition, the paper outlines the key performance metrics used in evaluating these detection mechanisms. Moreover, the novelty of the study is classifying the datasets commonly used for training and validating DDoS detection models into two major categories: legacy-compatible datasets that are adapted from traditional network environments, and SDN-contextual datasets that are specifically generated to reflect the characteristics of modern SDN systems. Finally, the paper suggests a few directions for future research. These include enhancing the robustness of detection models, integrating privacy-preserving techniques in collaborative learning, and developing more comprehensive and realistic SDN-specific datasets to improve the strength of SDN infrastructures against DDoS threats. Full article

In the realm of sustainable public transportation, the integration of intelligent electric bus propulsion systems represents a novel and promising approach to reducing environmental impact—particularly through the mitigation of NOx emissions and overall exhaust pollutants. This emerging technology underscores the growing need for advanced drive control architectures that ensure not only operational safety and reliability but also compliance with increasingly stringent emissions standards. The present article introduces an innovative analysis of energy-optimized dual-drive electric propulsion systems, with a specific focus on their potential for real-world application in emission-conscious urban mobility. A detailed dynamic model of a dual-drive electric bus was developed in MATLAB Simulink, incorporating a Fuzzy Logic-based decision-making algorithm embedded within the Transmission Control Unit (TCU). The proposed control architecture includes a torque-limiting safety strategy designed to prevent motor overspeed conditions, thereby enhancing both efficiency and mechanical integrity. Furthermore, the system architecture enables supervisory override of the Fuzzy Inference System (FIS) during critical scenarios, such as gear-shifting transitions, allowing adaptive control refinement. The study addresses the unique control and coordination challenges inherent in dual-drive systems, particularly in relation to optimizing gear selection for reduced energy consumption and emissions. Key areas of investigation include maximizing efficiency along the motor torque–speed characteristic, maintaining vehicular dynamic stability, and minimizing thermally induced performance degradation. The thermal modeling approach is grounded in integral formulations capturing major loss contributors including copper, iron, and mechanical losses while also evaluating convective heat transfer mechanisms to improve cooling effectiveness. These insights confirm that advanced thermal management is not only vital for performance optimization but also plays a central role in supporting long-term strategies for emission reduction and clean, efficient public transportation. Full article

Recent advances in 3D Gaussian Splatting (3DGS) have achieved remarkable performance in scene reconstruction and novel view synthesis on benchmark datasets. However, real-world images are frequently affected by degradations such as camera shake, object motion, and lens defocus, which not only compromise image quality but also severely hinder the accuracy of 3D reconstruction—particularly in fine details. While existing deblurring approaches have made progress, most are limited to addressing a single type of blur, rendering them inadequate for complex scenarios involving multiple blur sources and resolution degradation. To address these challenges, we propose Gaussian Splatting with Multi-Scale Deblurring and Resolution Enhancement (GS-MSDR), a novel framework that seamlessly integrates multi-scale deblurring and resolution enhancement. At its core, our Multi-scale Adaptive Attention Network (MAAN) fuses multi-scale features to enhance image information, while the Multi-modal Context Adapter (MCA) and adaptive spatial pooling modules further refine feature representation, facilitating the recovery of fine details in degraded regions. Additionally, our Hierarchical Progressive Kernel Optimization (HPKO) method mitigates ambiguity and ensures precise detail reconstruction through layer-wise optimization. Extensive experiments demonstrate that GS-MSDR consistently outperforms state-of-the-art methods under diverse degraded scenarios, achieving superior deblurring, accurate 3D reconstruction, and efficient rendering within the 3DGS framework. Full article

Additive manufacturing is driving a significant change in industry, extending beyond prototyping to the inclusion of printed parts in final designs. Stereolithography (SLA) is a polymerization technique valued for producing highly detailed parts with smooth surface finishes. This study presents a hybrid intelligent framework for modeling and optimizing the SLA 3D printer process’s parameters for Acrylonitrile Butadiene Styrene (ABS) photopolymer parts. The nonlinear relationships between the process’s parameters (Orientation, Lifting Speed, Lifting Distance, Exposure Time) and multiple performance characteristics (ultimate tensile strength, yield strength, modulus of elasticity, Shore D hardness, and surface roughness), which represent complex relationships, were investigated. A Taguchi design of the experiment with an L18 orthogonal array was employed as an efficient experimental design. A novel hybrid fuzzy logic–Particle Swarm Optimization (PSO) algorithm, ARGOS (Adaptive Rule Generation with Optimized Structure), was developed to automatically generate high-accuracy Mamdani-type fuzzy inference systems (FISs) from experimental data. The algorithm starts by customizing Modified Learn From Example (MLFE) to create an initial FIS. Subsequently, the generated FIS is tuned using PSO to develop and enhance predictive accuracy. The ARGOS models provided excellent performances, achieving correlation coefficients (R²) exceeding 0.9999 for all five output responses. Once the FISs were tuned, a multi-objective optimization was carried out based on the weighted sum method. This step helped to identify a well-balanced set of parameters that optimizes the key qualities of the printed parts, ensuring that the results are not just mathematically ideal, but also genuinely helpful for real-world manufacturing. The results showed that the proposed hybrid approach is a robust and highly accurate method for the modeling and multi-objective optimization of the SLA 3D process. Full article