Search Results (403)

Maritime UAV operations face challenges in real-time ship detection. Complex ocean backgrounds, drastic scale variations, and prevalent distant small targets create difficulties. We propose MCB-RT-DETR, a real-time detection transformer enhanced by multi-component boosting. This method builds upon the RT-DETR architecture. It significantly improves detection under wave interference, lighting changes, and scale differences. Key innovations address these challenges. An Orthogonal Channel Attention (Ortho) mechanism preserves high-frequency edge details in the backbone network. Receptive Field Attention Convolution (RFAConv) enhances robustness against background clutter. A Small Object Detail Enhancement Pyramid (SOD-EPN) strengthens small-target representation. SOD-EPN combines SPDConv with multi-scale CSP-OmniKernel transformations. The neck network integrates ultra-lightweight DySample upsampling. This enables content-aware sampling for precise multi-scale localization. The method maintains high computational efficiency. Experiments on the SeaDronesSee dataset show significant improvements. MCB-RT-DETR achieves 82.9% mAP@0.5 and 49.7% mAP@0.5:0.95. These correspond to improvements of 4.5% and 3.4% relative to the baseline model. Inference speed maintains 50 FPS for real-time processing. The outstanding performance in cross-dataset tests further validates the algorithm’s strong generalization capability on DIOR remote sensing images and VisDrone2019 aerial scenes. The method provides a reliable visual perception solution for autonomous maritime UAV operations. Full article

To enhance blockchain scalability, sharding technology enables parallel transaction processing, but existing solutions often neglect node heterogeneity, which introduces security risks and performance bottlenecks. This paper proposes a novel dynamic sharding scheme that dynamically allocates validators to shards based on their historical performance scores and computational power, ensuring balanced shard capacity and higher attack resistance. A tailored reward–penalty mechanism further incentivizes participation and discourages malicious behavior. Experimental evaluations demonstrate that our approach significantly outperforms prominent sharding protocols, including Elastico, OmniLedger, and RapidChain, by achieving higher throughput and lower latency. The proposed scheme effectively addresses node heterogeneity and enhances the overall scalability and security of blockchain systems. Full article

Background: Trabeculectomy and minimally invasive glaucoma surgery (MIGS) such as goniotomy aim to reduce intraocular pressure (IOP) and medication burden but are often performed in patients with differing disease severity. Methods: We retrospectively reviewed 100 eyes from 76 adults with glaucoma that underwent either goniotomy (n = 50; Kahook Dual Blade = 42, OMNI = 8) or trabeculectomy ab externo (n = 50) at a tertiary center between May 2022 and June 2023, with at least six months of follow-up. Baseline and six-month IOP, number of medications, and postoperative complications were recorded. Eyes undergoing trabeculectomy had higher preoperative IOP than those undergoing goniotomy (22.6 ± 7.7 vs. 19.1 ± 5.9 mmHg). Results: At six months, trabeculectomy achieved a greater absolute IOP reduction (8.8 ± 0.8 vs. 5.4 ± 0.8 mmHg; p = 0.004), likely reflecting higher baseline IOP, while goniotomy yielded a larger medication reduction (1.47 ± 0.30 vs. 0.72 ± 0.20; p = 0.041). Hyphema occurred more often after trabeculectomy, and the small number of OMNI cases precluded device comparison. Conclusions: In this short-term retrospective series, trabeculectomy achieved larger absolute IOP reduction whereas goniotomy offered greater medication reduction, highlighting the need to individualize surgical choice and confirm these findings in larger prospective studies. Full article

Spinach (Spinacia oleracea L.) is a commonly consumed crop with a diverse array of unique flavonoids. These molecules likely contribute to the health benefits associated with spinach consumption. However, little is known about the genetic diversity of these molecules, their bioaccessibility, and the heritability of these traits. We assembled a diversity panel of 30 F₁ and open-pollinated spinach accessions and cultivated them under controlled conditions over two periods. Quantification of 39 flavonoids revealed that their concentration is largely influenced by environmental factors, and at least two divergent branches in the spinach flavonoid biosynthesis pathway may exist. Despite generally similar trends in the amounts of major flavonoids, open-pollinated and F₁ varieties of spinach could be distinguished based on the concentrations of minor flavonoid species. Broad-sense heritability estimates for absolute bioaccessibility accounted for more genetic variation than raw material content, suggesting that this trait is preferable for breeders seeking to alter the phytochemical profile of spinach. Lastly, we found that several spinach flavonoids are unstable under digestive conditions, which was made evident by the proportion of aglycones rising from 0.1% to approximately 15% of total flavonoids after digestion. Together, these data suggest that spinach flavonoid biosynthesis and bioaccessibility are complex and contextualize how these molecules may behave in vivo. Full article

Objectives: To examine three elastic band resistance training (EB-RT) modalities—high-velocity (HVRT), accentuated eccentric (Aecc), and maximal strength (Max)—on bone health, strength, redox-inflammatory profile, and neuroplasticity in sedentary older adults. Methods: Sixty-one participants (69.41 ± 4.61 years) were randomly assigned to HVRT (n = 21), Aecc (n = 13), Max (n = 10), or passive controls (n = 17). Training was conducted three times a week for 16 weeks. Sessions included four sets of alternating upper- and lower-limb EB exercises, with intensity guided by the OMNI–RES EB scale. HVRT emphasized explosive concentric actions [~70% one-repetition maximum (1RM); 3–4 rating of perceived exertion in the first repetition (RPE-1)]. Aecc performed 5 s eccentric overload [>100% 1RM; 7–8 RPE-1]. Max employed controlled 2 s concentric/eccentric actions [~80–85% 1RM; 7–8 RPE-1]. Results: All training groups improved isokinetic strength (p < 0.01, g = 0.91–2.40). HVRT increased brain-derived neurotrophic factor (BDNF) (p = 0.019, g = 0.42) and glutathione peroxidase (GPx) (p < 0.001, g = 0.31). Aecc elicited the strongest osteoanabolic and antioxidant effects (P1NP, p = 0.001, g = 1.21; β-CTX, p < 0.001, g = 1.82; F2-isoprostanes, p = 0.007, g = 0.94). Max induced moderate bone turnover benefits (P1NP, p = 0.005, g = 1.08; β-CTX, p < 0.001, g = 1.12), but no GPx or BDNF gains. Controls maintained or declined all variables. Conclusions: EB-RT over 16 weeks improved most outcomes overall, showing modality-specific trends: HVRT favored neuroplasticity, Aecc enhanced redox-inflammatory and bone remodeling responses, and Max improved strength and bone health. These findings support elastic band resistance training as a safe and individualized strategy for healthy aging. Full article

Uniform spraying by conventional plant protection drones often results in low herbicide utilization efficiency and environmental contamination, both of which are critical issues in agricultural production. To address these challenges, this study proposed a precision weed management system for maize fields that combines an improved YOLOv11n-OSAW detection model with DJI drones for variable-rate herbicide application. The YOLOv11n-OSAW model was enhanced with Omni-dimensional Dynamic Convolution (OD-Conv), the SEAM attention mechanism, a lightweight ADown module, and the Wise-IoU (WIoU) loss function, aiming to improve the detection accuracy of small and occluded weeds in maize fields. When the model was deployed on an uncrewed aerial vehicle (UAV) operating at 5 m altitude, it achieved mean Average Precision mAP@0.5 values of 97.8% and 97.0% for gramineous and broad-leaved weeds, respectively—representing increases of 2.9 and 1.6 percentage points over the baseline YOLOv11n model. Weed distribution maps generated from the detection results were used to develop site-specific herbicide prescription maps, guiding the drone to implement targeted spraying. Water-sensitive paper analysis verified that the system ensured effective droplet deposition and uniform coverage across different application rate areas. This integrated workflow, covering UAV image acquisition, weed detection, variable-rate application, and effect assessment, reduced herbicide consumption by 20.25% compared with conventional uniform spraying (450 L/ha) while maintaining excellent weed control efficiency and reducing environmental risks. The findings demonstrate that the proposed system provides a practical and sustainable solution for weed management in maize fields. Full article

With rising demand for cross-camera person re-identification (ReID) in smart cities, low-resolution (LR) images severely hinder practical ReID performance due to detail loss and weakened identity features. This paper proposes two solutions to address this bottleneck: (1) super-resolution (SR) techniques, including hybrid attention transformer (HAT), pixel-level and semantic-level adjustable SR (PiSA-SR), and omni aggregation networks for lightweight image SR (Omni-SR), are used to enhance image visual quality, and the enhanced images are applied to three ReID methods, including semantically controllable self-supervised learning framework-REID (SOLIDER-REID), light-REID, and relation-aware global attention (RGA), for performance assessment. (2) An end-to-end framework integrating HAT and SOLIDER-REID is designed, in which HAT enhances LR images via multi-scale attention to restore discriminative details, while SOLIDER-REID’s semantic controller suppresses background noise to focus on the pedestrian regions. Extensive experiments on the Market-1501 dataset show that the first solution slightly improves ReID accuracy, e.g., PiSA-SR + SOLIDER-REID achieves 92.0% mAP, 0.4% higher than SOLIDER-REID alone, while slightly sacrificing speed. The second solution significantly boosts LR ReID performance at the cost of a certain increase in time. For LR images, even 32 × 32 images, HAT-SOLIDER achieves 59.8% mAP and 80.4% Rank-1, 18.5% higher in mAP and 19.2% higher in Rank-1 than SOLIDER-REID alone. This work provides effective solutions for LR-induced performance degradation in cross-camera ReID. Full article

Cardiopulmonary resuscitation (CPR) is a vital skill for healthcare professionals, crucial in life-saving situations. More than 80% of cardiac arrest cases occur out of hospital. As the demand for competent CPR practitioners grows, the effectiveness of training methods becomes increasingly important, especially for undergraduate students preparing to enter the healthcare field. The primary objective of our study is to investigate the effectiveness of simulation-based teaching methods and by integrating innovative technologies, such as the OMNI2 simulator, to enhance practitioners’ performance and to improve the precision and objectivity of CPR instruction. A cohort of 144 undergraduate students from the Nursing School Department of the National Kapodistrian University of Athens participated in an 8 h Basic Life Support Seminar. It consisted of a 5 h theoretical instruction followed by 3 h of practical training using the OMNI2 simulator. Each student was tasked to identify cardiac arrest and to perform two cycles of CPR according to the 2021 guidelines. Metrics, including total session time, cycles performed, compression-to-ventilation ratio, compression depth, compressions and ventilations per minute, full recoil, peak inspiratory pressure, and ventilation duration, were measured and compared against the simulator’s preset targets. Statistically significant differences (p < 0.05) were observed for all outcomes. In conclusion, while simulation-based teaching has conventionally been proven effective for CPR proficiency, real-time data collected in this study reveal a disparity between anticipated and actual performance. Our research underscores the necessity of refining instructional methods to enhance skill acquisition, potentially leading to improved patient outcomes in the future. Full article

Automatic speech recognition (ASR) technology faces the dual challenges of model complexity and noise robustness when deployed on terminal devices (e.g., mobile devices, embedded systems). To meet the demand for lightweight and high-performance models in terminal devices, we propose a lightweight end-to-end speech recognition model, OS-Denseformer (Omni-Scale-Denseformer). The core of this model lies in its lightweight design and noise adaptability: multi-scale acoustic features are efficiently extracted through a multi-sampling structure to enhance noise robustness; the proposed OS-Conv module improves local feature extraction capability while significantly reducing the number of parameters, enhancing computational efficiency, and lowering model complexity; the proposed normalization function, ExpNorm, normalizes the model output, facilitating more accurate parameter optimization during model training. Finally, we employ distinct loss functions across different training stages, using Minimum Bayes Risk (MBR) joint optimization to determine the optimal weighting scheme that directly minimizes the character error rate (CER). Experimental results on public datasets such as AISHELL-1 demonstrate that, under a high-noise environment of −15 dB, the CER of the OS-Denseformer model is reduced by 9.95%, 7.97%, and 4.85% compared to the benchmark models Squeezeformer, Conformer, and Zipformer, respectively. Additionally, the model parameter count is reduced by 53.35%, 10.27%, and 27.66%, while the giga floating-point operations per second (GFLOPs) are decreased by 67.51%, 66.51%, and 13.82%, respectively. Deployment on resource-constrained mobile devices demonstrates that, compared to Conformer, OS-Denseformer reduced memory usage by 10.79% and decreased inference latency by 61.62%. Full article

Amid the urgent demands for global trade transformation and zero-carbon transition, sustainable maritime supply chains face challenges of high costs and complex coordination, necessitating the elimination of “isolated decision-making” to achieve sustainable development goals. This study constructs a profit analysis model under centralized and decentralized decision-making scenarios and various intelligent omni-channel models, exploring the profit composition, optimal pricing, and operational strategies of carriers and forwarders. Case analysis validates that collaborative optimization, particularly when forwarders leverage online channels and customer proximity, enables sustainable maritime transport and significantly enhances overall profits and efficiency in sustainable maritime supply chains. This research provides a theoretical and practical framework for collaborative optimization strategies contributing to sustainable maritime transport and port intelligence in marine engineering contexts. By this framework, it will be possible to advance green transformation, smart operations management, and digital innovation in the global maritime and marine industries. Full article

Based on the self-determination theory and social presence theory, this study examined how high-tech and high-touch orientations in logistics service quality (LSQ) influence consumer satisfaction in omni-channel retailing. LSQ was modelled as two second-order constructs: high-tech orientation (timeliness, physical facilities, and ease of return) and high-touch orientation (employees’ knowledge, flexibility, and responsiveness to delivery discrepancies). Survey data from 455 consumers were analyzed using structural equation modelling. Both orientations significantly improved satisfaction, with high-tech orientation showing a slightly stronger effect, reflecting the digital literacy of the predominantly young sample. The findings extended self-determination theory and social presence theory by offering a dual-orientation perspective and practical guidance for balancing high-tech and high-touch in omni-channel logistics service design. Full article

Nutritional and cooling strategies to abate the negative effects of heat stress during the dry period have been used to improve the performance of dairy cattle. The objective of this study was to evaluate the effects of feeding an immunomodulatory supplement (OmniGen-AF^®, OMN) before, during, and after exposure to either heat stress or active cooling during the dry period on immune function and mammary development in dairy cows. During late lactation (at least 60 d before dry off), cows were provided with evaporative cooling systems (shade, fans, and soakers) and assigned to two groups: placebo (56 g/d of AB20^® top-dressed; CON) or OmniGen-AF^® (OMN, 56 g/d top-dressed). Cows were dried off ~46 d before the expected calving date and further split into evaporative cooling (shade, fans, and soakers; CL) or heat stress (only shade; HT) pens. Thus, after dry off, there were four treatment groups: heat stress with placebo (HT, n = 17), HT with OMN supplementation (HT + OMN, n = 19), CL with placebo (CL, n = 16), and CL with OMN supplementation (CL + OMN, n = 11). From a subset of cows (n = 6–8 per group), four blood samples were collected during the dry period (−43, −39, −32, and −21 d relative to calving) to evaluate neutrophil function and blood hematology. In addition, mammary biopsies (4–6 cows/treatment) were collected at −43, −39, −32, and −21 d relative to calving to evaluate mammary gland gene expression and histology, i.e., Tdt dUTP nick-end labeling (TUNEL) and Ki67. Genes related to autophagy, apoptosis, and cell proliferation were analyzed by qRT-PCR. Relative to CL, HT downregulated the expression of beclin-2 (BECN2) but upregulated the expression of beclin-1 (BECN1) on days −43 and −39 relative to calving, respectively. Also, relative to CL, HT upregulated the expression of BAX and FAS on day −39 relative to calving. These differences in gene expression were followed by HT cows having a lower total cell apoptosis rate during involution relative to CL cows. Further to these effects, HT leads to a lower alveoli number relative to CL cows. As in the CL treatment, OMN cows have a higher total cell apoptosis rate and alveoli number relative to CON cows. In addition, OMN cows have higher total cell proliferation relative to CON. Prolactin (PRL) and cortisol concentrations were evaluated during the dry period at days −45, −26, −3, and −1 relative to calving. Relative to CL, HT cows had higher PRL at day −45 but lower PRL on day −1 relative to calving, and a similar trend was observed for cortisol concentrations. In summary, HT impacts mammary gland gene expression, compromises mammary involution, reduces alveoli number, and alters hormone dynamics throughout the dry period. Following the same trends as the CL treatment, OMN increases mammary gland turnover by having a higher cell apoptosis and cell proliferation rate and lower connective tissue relative to CON cows. Full article

Blockchain technology, originally designed as a secure and immutable ledger, has expanded its applications across various domains. However, its scalability remains a fundamental bottleneck, limiting throughput, specifically Transactions Per Second (TPS) and increasing confirmation latency. Among the many proposed solutions, sharding has emerged as a promising Layer 1 approach by partitioning blockchain networks into smaller, parallelized components, significantly enhancing processing efficiency while maintaining decentralization and security. In this paper, we have conducted a systematic literature review, resulting in a comprehensive review of sharding. We provide a detailed comparative analysis of various sharding approaches and emerging AI-assisted sharding approaches, assessing their effectiveness in improving TPS and reducing latency. Notably, our review is the first to incorporate and examine the standardization efforts of the ITU-T and ETSI, with a particular focus on activities related to blockchain sharding. Integrating these standardization activities allows us to bridge the gap between academic research and practical standardization in blockchain sharding, thereby enhancing the relevance and applicability of our review. Additionally, we highlight the existing research gaps, discuss critical challenges such as security risks and inter-shard communication inefficiencies, and provide insightful future research directions. Our work serves as a foundational reference for researchers and practitioners aiming to optimize blockchain scalability through sharding, contributing to the development of more efficient, secure, and high-performance decentralized networks. Our comparative synthesis further highlights that while Bitcoin and Ethereum remain limited to 7–15 TPS with long confirmation delays, sharding-based systems such as Elastico and OmniLedger have reported significant throughput improvements, demonstrating sharding’s clear advantage over traditional Layer 1 enhancements. In contrast to other state-of-the-art scalability techniques such as block size modification, consensus optimization, and DAG-based architectures, sharding consistently achieves higher transaction throughput and lower latency, indicating its position as one of the most effective Layer 1 solutions for improving blockchain scalability. Full article

Accurate pedestrian detection from a robotic perspective has become increasingly critical, especially in complex environments such as crowded and high-density populations. Existing methods have low accuracy due to multi-scale pedestrians and dense occlusion in complex environments. To address the above drawbacks, a dense pedestrian detection network architecture based on YOLOv8n (DPDN-YOLOv8) was introduced for complex environments. The network aims to improve robots’ pedestrian detection in complex environments. Firstly, the C2f modules in the backbone network are replaced with C2f_ODConv modules integrating omni-dimensional dynamic convolution (ODConv) to enable the model’s multi-dimensional feature focusing on detected targets. Secondly, the up-sampling operator Content-Aware Reassembly of Features (CARAFE) is presented to replace the Up-Sample module to reduce the loss of the up-sampling information. Then, the Adaptive Spatial Feature Fusion detector head with four detector heads (ASFF-4) was introduced to enhance the system’s ability to detect small targets. Finally, to accelerate the convergence of the network, the Focaler-Shape-IoU is utilized to become the bounding box regression loss function. The experimental results show that, compared with YOLOv8n, the mAP@0.5 of DPDN-YOLOv8 increases from 80.5% to 85.6%. Although model parameters increase from $3\times {10}^6$ to $5.2\times {10}^6$, it can still meet requirements for deployment on mobile devices. Full article

Single-cell RNA sequencing (scRNA-seq) has revolutionized genomic investigations by enabling the exploration of gene expression heterogeneity at the individual cell level. However, the complexity of scRNA-seq data analysis remains a challenge for many researchers. Here, we present OmniCellX, a browser-based tool designed to simplify and streamline scRNA-seq data analysis while addressing key challenges in accessibility, scalability, and usability. OmniCellX features a Docker-based installation, minimizing technical barriers and ensuring rapid deployment on local machines or clusters. Its dual-mode operation (analysis and visualization) integrates a comprehensive suite of analytical tools for tasks such as preprocessing, dimensionality reduction, clustering, differential expression, functional enrichment, cell–cell communication, and trajectory inference on raw data while enabling alternative interactive and publication-quality visualizations on pre-analyzed data. Supporting multiple input formats and leveraging the memory-efficient data structure for scalability, OmniCellX can efficiently handle datasets spanning millions of cells. The platform emphasizes user flexibility, offering adjustable parameters for real-time fine-tuning, alongside extensive documentation to guide users at even beginner levels. OmniCellX combines an intuitive interface with robust analytical power to perform single-cell data analysis and empower researchers to uncover biological insights with ease. Its scalability and versatility make it a valuable tool for advancing discoveries in cellular heterogeneity and biomedical research. Full article