Search Results (21,021)

This study presents a comparative analysis of hyperspectral image segmentation algorithms for fruit defect detection under different illumination conditions. The research evaluates the performance of four segmentation methods (Spectral Angle Mapper, Random Forest, Support Vector Machine, and Neural Network) using three distinct illumination modes (local, simultaneous and sequential). The experimental setup employed hyperspectral imaging to assess tomato fruit samples, with data acquisition performed across the 450–850 nm spectral range. Quantitative metrics, including accuracy, error rate, precision, recall, F1-score, and Intersection over Union (IoU), were used to evaluate algorithm performance. Key findings indicate that Random Forest demonstrated superior performance across most metrics, particularly under simultaneous illumination conditions. The highest accuracy was achieved by Random Forest under sequential illumination (0.9971), while the best combination of segmentation metrics was obtained under simultaneous illumination, with an F1-score of 0.8996 and an IoU of 0.8176. The Neural Network showed competitive results. The Spectral Angle Mapper proved sensitive to illumination variations but excelled in specific scenarios requiring minimal memory usage. By demonstrating that acquisition protocol optimization can substantially improve segmentation performance, our results support the development of accurate, non-contact, high-throughput inspection systems and contribute to reducing postharvest losses and improving supply chain quality control. Full article

Traffic variability along heavily congested signalised urban corridors undermines roadway safety, reduces energy efficiency, weakens operational reliability, and can hinder emergency response. Although many simulation-based studies have examined the impacts of Autonomous Vehicles (AVs), relatively few have combined high-resolution congestion observations with link-level microscopic calibration in a real urban network, particularly when evaluating implications for emergency mobility. This study develops and calibrates a microscopic Aimsun traffic simulation model for the Atakum district of Samsun, Türkiye, using a 10 min Google Traffic Index (GTI) observation stream converted into a four-level ordinal congestion scale. The calibration process began with an origin–destination (OD) matrix derived from 2020 traffic counts and was refined through link-level GTI synchronization, iterative OD scaling on mismatched corridors, and signal retiming at key intersections. GTI was validated as an ordinal congestion proxy through both categorical agreement and volumetric consistency, achieving 83% class agreement and GEH values below 5 for more than 90% of links. Five AV penetration scenarios (0%, 25%, 50%, 75%, and 100%) were simulated under peak-hour conditions. Network performance was evaluated using delay, stop time, mean speed, throughput, missed turns, and total journey time, while emergency mobility was assessed along a representative ambulance corridor on Atatürk Boulevard using seconds per kilometre. The results indicate that increasing AV penetration improves flow stability more clearly than nominal capacity. Mean speed increased from 36.2 to 39.2 km/h, delay and stop time declined steadily, and throughput remained nearly constant at 22.2–22.5 thousand vehicles/h. Along the ambulance corridor, travel time improved by 11.5%, from 112.4 to 99.4 s/km, between the baseline and full automation scenarios. These findings provide scenario-based evidence that, within a calibrated signalised urban network, increasing AV penetration can enhance operational stability and emergency response efficiency. More broadly, the study demonstrates the practical value of integrating GTI-based congestion observations with microscopic simulation for AV impact assessment in real urban networks. Full article

Demand for camel milk products is growing in Morocco and worldwide, creating opportunities to strengthen the livelihoods of populations living in arid regions through the development of camel-based dairy value chains. In addition to their economic potential, such value chains may contribute to sustainability by supporting food systems adapted to arid environments, promoting the use of locally resilient livestock species, and enhancing the socio-economic viability of vulnerable rural communities. This exploratory qualitative study investigates urban consumer behavior related to dairy consumption with a specific focus on the potential integration of camel milk products into local dietary habits. To capture nuanced consumer perspectives, gender-segregated focus-group discussions were conducted in three Moroccan cities using a semi-structured questionnaire on dairy consumption habits. Key factors examined included milk types, product preferences, purchasing locations, consumption frequency and willingness to include camel products in the household diet. The results indicate that camel milk is rarely consumed outside areas where camels are raised. Nevertheless, participants expressed interest in several camel milk-based products, particularly fermented milk and spreadable cheeses. This interest was primarily driven by perceptions of camel milk as a healthy product and by its association with traditional food practices. These findings suggest that expanding camel milk consumption in urban markets could support more sustainable and territorially rooted dairy systems by linking consumer demand with production models suited to dryland conditions. This study indicates promising market opportunities for the development of camel milk products in urban areas, particularly if challenges related to pricing strategies, distribution network, and region-specific supply chains are strategically managed. Full article

Soil salinization is a major threat to global crop production. Tartary buckwheat (Fagopyrum tataricum), valued for its hardiness in marginal environments, provides an excellent system for studying plant salt tolerance. Using an integrated multi-omics approach, we deciphered the physiological, metabolic, and transcriptional responses of Tartary buckwheat to prolonged NaCl stress. Physiological profiling confirmed membrane damage alongside osmotic adjustment via proline accumulation and a phased antioxidant response. Metabolomics revealed extensive reprogramming, with dynamic enrichment in pathways of flavonoid biosynthesis, lipid metabolism, and the TCA cycle. Transcriptomics delineated a time-specific cascade from early signaling to late defense activation. Critical regulators within ABA and MAPK signaling pathways showed fine-tuned, divergent expression; for instance, SnRK2.3 was suppressed while specific PP2Cs were induced, and FtMAPK10 was dramatically up-regulated. Integrated analysis demonstrated coordinated induction of osmoprotectant synthesis (e.g., galactinol and betaine pathways) and a rewiring of central carbon metabolism. Our findings reveal a sophisticated, multi-layered adaptation strategy in Tartary buckwheat, integrating enhanced osmolyte production, antioxidant defense, membrane remodeling, and metabolic reprogramming, orchestrated by key signaling networks. This study provides a comprehensive molecular framework for salt tolerance and identifies valuable genetic targets for improving crop resilience. Full article

Accurate prediction of a ship’s Estimated Time of Arrival (ETA) is of great significance for port scheduling, logistics management, and navigation safety. Traditional ETA prediction approaches often rely on manual experience for parameter tuning, which tends to be inefficient and susceptible to subjective factors. To address this issue and improve prediction accuracy, this study proposes a hybrid modeling framework, integrating Bayesian Optimization (BO), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks. In this approach, Automatic Identification System (AIS) data is leveraged to predict the total voyage duration before departure, thereby deriving the vessel’s ETA. The model, referred to as BO-CNN-LSTM, utilizes BO for automatic hyperparameter tuning, employs CNN for extracting local features, and applies LSTM network to capture temporal dependencies. The model is developed using a dataset of 32,972 distinct voyage records, among which 23,947 are retained as valid samples after data cleaning. Pearson correlation analysis is conducted to select key input variables, including navigation speed, ship type, sailing distance, and deadweight tonnage. Additionally, sailing distance is processed using the Ramer–Douglas–Peucker algorithm. Experimental evaluation indicates that the BO-CNN-LSTM model achieves a coefficient of determination of 0.987, along with a mean absolute error and root mean square error of 6.078 and 8.730, respectively. These results significantly outperform comparison models such as CNN, LSTM, CNN-LSTM, random forest, AdaBoost, and Elman neural networks. Overall, this study validates the effectiveness and superiority of the proposed BO-CNN-LSTM model in ship ETA prediction, providing an efficient and effective prediction solution for intelligent maritime transportation systems. Full article

With the continuous development of the petroleum industry, bottomhole pressure prediction technology, which exerts a significant impact on oil production and recovery, has become a key research direction in the current oil and gas field. To enhance the accuracy and robustness of bottomhole pressure prediction under transient and variable operating conditions, a method based on data augmentation strategies and hyperparameter optimization was proposed in this paper. Addressing challenges such as limited data volume and significant disturbances in actual oilfield production, a data augmentation strategy incorporating noise perturbation and sliding windows was introduced to expand training samples and improve model generalization. In terms of model architecture, a deep network integrating CNN, BiGRU, and Multi-Head Attention mechanisms was proposed in this paper, which is referred to as the CNN-BiGRU-Multi-Head Attention model. By introducing Bayesian optimization for automatic hyperparameter search, the performance of the temporal model was further enhanced, achieving efficient extraction and dynamic focusing of wellbore pressure temporal features. Prediction results demonstrated that the proposed method outperforms existing mainstream forecasting models in metrics such as Mean Absolute Error (MAE) and Coefficient of Determination (R²), with R² reaching 0.9831, which confirms its strong generalization capability and engineering applicability. Practical guidance for intelligent oilfield production management and bottomhole pressure forecasting, along with a novel prediction method, is provided by this study, which holds significant importance for extending well life and stabilizing hydrocarbon production. Full article

Inertial measurement units (IMUs) in low-cost navigation systems suffer from significant drift and noise errors due to sensor biases, scale factor instability, and nonlinear stochastic noise. This paper proposes a hybrid error compensation approach that combines Fast Orthogonal Search (FOS), a nonlinear system identification technique, with deep Long Short-Term Memory (LSTM) neural networks to improve IMU signal accuracy in GNSS-denied navigation. The FOS algorithm efficiently models deterministic error patterns (such as bias drift and scale factor errors) using a small training dataset, while the LSTM learns the IMU’s complex time-dependent error dynamics from much longer training data. In the proposed method, FOS is first used to predict the output of a high-end IMU based on that of a low-end IMU, and the trained FOS model is then used to extend the training data for an LSTM-based predictor. We demonstrate the efficacy of this FOS–LSTM hybrid on real vehicular IMU data by training with a limited segment of high-precision reference measurements and testing on extended operation periods. The hybrid model achieves high predictive accuracy for predicting the high-end signal based on the low-end signal, with a mean squared error below 0.1% and yields more stable velocity estimates than models using FOS or LSTM alone. Although long-term position drift is not fully eliminated, the proposed method significantly reduces short-term uncertainty in the inertial solution. These results highlight a promising synergy between model-based system identification and data-driven learning for sensor error calibration in navigation systems. Key contributions include FOS-based pseudo-label bootstrapping for data-efficient LSTM training and a navigation-level evaluation illustrating how signal correction impacts dead reckoning drift. Full article

The industrial metaverse is a digital twin space that integrates the real world with virtual environments through bidirectional synchronization. It supports critical services, such as time-sensitive machine control and large-scale collaboration, which require Time-Sensitive Networking and scalable Data Distribution Services. DDS, developed by the Object Management Group, provides excellent scalability and diverse QoS policies but struggles to guarantee transmission delay and jitter for time-critical applications. TSN, based on IEEE 802.1 standards, addresses these challenges by ensuring time-criticality. However, current research lacks comprehensive integration mechanisms for DDS and TSN, particularly from the viewpoints of semantics and system framework. Additionally, there is no adaptive QoS mapping converting the abstract DDS QoS policies to the sophisticated TSN QoS parameters. This paper presents a novel DDS-over-TSN framework that incorporates three key functions to address these challenges. First, Cross-layer QoS Mapping automates correspondences between DDS and TSN parameters, deriving technical constraints from standard documentation through retrieval-augmented generation. Second, Semantic Priority Estimation extracts substantial priority levels by utilizing language model embedding vectors as high-dimensional feature extractors. Third, Adaptive Resource Allocation performs dynamic bandwidth distribution for each priority level through reinforcement learning. Simulation results reveal over 99% mapping accuracy and 97% consistency in priority extraction. The applied Deep Reinforcement Learning paradigm allocated 99% of required resources to high-priority classes and reduced resource wastage by 15% compared to conventional methods. This methodology meets industrial requirements by ensuring both deterministic real-time performance and efficient resource isolation. Full article

Pedestrian gender recognition plays an important role in pedestrian analysis and intelligent video applications, for example, in demographic statistics, soft biometric analysis, and context-aware person retrieval. However, it remains a challenging task owing to viewpoint variations, illumination changes, occlusions, and low image quality in real-world imagery. To address these issues, an effective adaptive decision fusion framework, termed the Decision Fusion Learning Network (DFLN), is proposed in this paper. The key novel aspect of DFLN is that it effectively explores both an appearance-centered view that emphasizes detailed texture and clothing information and a structure-centered view that captures rich contour and structural information for pedestrian gender recognition. To realize DFLN, a Parallel CNN Prediction Probability Learning Module (PCNNM) is first constructed to independently learn modality-specific probabilities from color image and edge maps. Subsequently, a learnable Decision Fusion Module (DFM) is designed to fuse the modality-specific probabilities and explore their complementary merits for realizing accurate pedestrian gender recognition. The DFM can be easily coupled with the PCNNM, forming an end-to-end decision fusion learning framework that simultaneously learns the feature representations and carries out adaptive decision fusion. Experiments on two pedestrian benchmark datasets, named PETA and PA-100K, show that DFLN achieves competitive or superior performance compared with several state-of-the-art pedestrian gender recognition methods. Extensive experimental analysis further confirms the effectiveness of the proposed decision fusion strategy and its favorable generalization ability under domain shift. Full article

Traditional journalists now utilize social media platforms such as Facebook, Instagram, Twitter, and TikTok to disseminate information. With the emergence of TikTok as a prominent social network for entertainment and information, many journalists worldwide, including in Israel, have begun leveraging it to create and share short video content. This study presents a qualitative case study of journalists operating within the Israeli media system, examining why and how journalists use TikTok, the professional challenges they face on the platform, and how they address these challenges. Specifically, it focuses on how journalists perceive TikTok as a journalistic space and their professional role within it. Focusing on the Israeli context, which is both digitally advanced and characterized by a democratic and pluralistic media environment, semi-structured in-depth interviews were conducted with 15 prominent journalists from traditional Israeli media outlets who are extensively active and considered at least micro-influencers on TikTok. The findings reveal several key themes regarding journalists’ use of TikTok. These include the platform’s role as a tool for reaching younger audiences and maintaining relevance; and the journalists’ self-perception as gatekeepers combating fake news. However it was found that they face ethical dilemmas and an absence of the structural and ethical foundations necessary for serious investigative journalism. This is the result of adapting their work to the platform’s light, fast-paced, and visually engaging format, favoring content that is entertaining and often sensational, to meet the expectations of TikTok audiences. While grounded in the Israeli case, the findings contribute to broader discussions on the platformization of journalism and the transformation of professional norms in media environments. Full article

Alcohol-induced liver damage (AILD), characterized by oxidative stress and inflammation, is a major health concern. While black ginseng extract (BGE) exhibits diverse pharmacological activities, its protective effects against AILD and underlying molecular mechanisms remain unclear. This study evaluated the protective effects of BGE against AILD using in vivo, in vitro, and in silico models. In mice, daily oral administration of 25% ethanol (5 g/kg) for 2 weeks induced liver injury. BGE (100–500 mg/kg) significantly reduced serum alanine aminotransferase (AST) and aspartate aminotransferase (ALT)levels while increasing catalase (CAT) and superoxide dismutase (SOD) activities. In ethanol-treated HepG2 cells, BGE inhibited nitric oxide (NO) production and suppressed cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6) expression while increasing heme oxygenase-1 (HO-1)expression. Ginsenoside Rh1, quantified at 4.7 mg/g via quadrupole linear ion trap tandem mass spectrometry coupled with UPLC (UPLC-Q-TRAP-MS/MS), was identified as a key bioactive compound. Network pharmacology and molecular docking analyses revealed key inflammatory signaling pathways and core hub genes associated with ginsenoside Rh1. Integrated analyses suggest that ginsenoside Rh1 contributes to the multi-target effects of BGE by modulating inflammatory signaling pathways. Collectively, BGE is a potential therapeutic candidate for the prevention and treatment of AILD, with ginsenoside Rh1 serving as a key bioactive constituent and quality control marker. Full article

Brain-Derived Neurotrophic Factor (BDNF), a key member of the neurotrophin family, is critically involved in neuronal survival, synaptic plasticity, learning, and memory. While its roles in mammals have been extensively documented, the molecular regulatory mechanisms governing BDNF expression and its causal contributions to complex cognitive behaviors remain poorly understood in non-mammalian vertebrates—particularly for the domestic pigeon (Columba livia domestica), a species distinguished by its remarkable spatial navigation and homing capabilities. This review synthesizes the current evidence on BDNF in the pigeon central nervous system across five thematic domains: molecular structure and isoform diversity, transcriptional and epigenetic regulatory networks, involvement in neural development, associations with cognitive and navigational behaviors, and potential translational applications. A particular emphasis is placed on the region-specific and activity-dependent expression patterns of BDNF in brain structures such as the hippocampal formation (HF), optic tectum, and striatum, and their functional relevance to visual processing, homing behavior, and stress adaptation. To date, most findings remain correlational; therefore, establishing a mechanistic understanding necessitates the integration of advanced methodologies—including single-cell omics, CRISPR-based gene editing, and high-resolution behavioral phenotyping—to causally link BDNF dynamics, neural circuit modulation, and spatial cognition. This synthesis aims to bridge gaps in comparative neurobiology, inform molecular approaches to avian cognitive enhancement, and support evidence-based strategies for racing pigeon breeding and welfare assessment. Full article

Fruit volatile organic compounds (VOCs) are key determinants of plum flavor quality, and esters contribute strongly to the fruity aroma of ripe fruit. However, the molecular basis of cultivar differences in ester formation during ripening has not been systematically clarified. Here, we characterized pulp VOC profiles across ripening in three Chinese plum (Prunus salicina) cultivars (‘WeiWang’ (WW), ‘WeiDi’ (WD), and ‘KongLongDan’ (KLD)) and integrated transcriptome analysis with weighted gene co-expression network analysis (WGCNA) to identify genes associated with ester accumulation. HS-SPME-GC-MS identified 38 VOCs, mainly esters, aldehydes, and alcohols, with ‘WW’ showing the highest total VOC abundance. During ripening, esters became the predominant volatile class in ‘WW’ and ‘WD’, in agreement with their fruity sensory characteristics, whereas ‘KLD’ maintained a more balanced composition of fruity and green-related volatiles. Transcriptomic analyses highlighted Prunus salicina alcohol acyltransferase 3 (PsAAT3) as the most abundant AAT transcript in pulp and strongly induced in ‘WW’. Transient overexpression of PsAAT3 in the low-ester background increased butyl acetate and hexyl acetate by 4.8- and 2.2-fold, respectively. WGCNA further identified ester-associated modules and candidate transcription factors co-expressed with PsAAT3 (JA2L, HY5, NAC073, and PHL13). As a result, this study identifies PsAAT3 as a key determinant of high-ester aroma in Chinese plum and provide candidate targets for aroma improvement and flavor-oriented breeding. Full article

Radiation pneumonitis (RP), one of the most common and severe complications in locally advanced non-small cell lung cancer (LA-NSCLC) patients following thoracic radiotherapy, presents significant challenges in prediction due to the complexity of clinical risk factors, incomplete multimodal data, and unavailable slice-level annotations in pre-radiotherapy CT images. To address these challenges, we propose a multimodal fusion framework based on Dual-Layer Attention-Based Adaptive Bag Embedding Multiple-Instance Learning (DAAE-MIL) for accurate RP prediction. This study retrospectively collected data from 995 LA-NSCLC patients who received thoracic radiotherapy between November 2018 and April 2025. After screening, Subject datasets (n = 670) were allocated for training (n = 535), and the remaining samples (n = 135) were reserved for an independent test set. The proposed framework first extracts pre-radiotherapy CT image features using a fine-tuned C3D network, followed by the DAAE-MIL module to screen critical instances and generate bag-level representations, thereby enhancing the accuracy of deep feature extraction. Subsequently, clinical data, radiomics features, and CT-derived deep features are integrated to construct a multimodal prediction model. The proposed model demonstrates promising RP prediction performance across multiple evaluation metrics, outperforming both state-of-the-art and unimodal RP prediction approaches. On the test set, it achieves an accuracy (ACC) of 0.93 and an area under the curve (AUC) of 0.97. This study validates that the proposed method effectively addresses the limitations of single-modal prediction and the unknown key features in pre-radiotherapy CT images while providing significant clinical value for RP risk assessment. Full article

Short-term load forecasting plays a pivotal role in modern power system operations yet it remains challenging due to the complex spatiotemporal dependencies in load data. This paper proposes a dual-head attention residual network (DARNet) that significantly advances STLF through three key innovations: (1) a hybrid encoder combining 1D-CNN and GRU architectures to simultaneously capture the local load patterns and long-term temporal dependencies, achieving a 28% better locality awareness than that of conventional approaches; (2) a novel dual-head attention mechanism that dynamically models both the inter-temporal relationships and cross-variable dependencies, reducing the feature engineering requirements; and (3) an autocorrelation-adjusted recursive forecasting framework that cuts the multi-step prediction error accumulation by 33% compared to that with standard seq2seq models. Extensive experiments on real-world datasets from three Chinese cities demonstrate DARNet’s superior performance, outperforming six state-of-the-art benchmarks by 21–35% across all of the evaluation metrics (MAPE, SMAPE, MAE, and RRSE) while maintaining robust generalization across different geographical regions and prediction horizons. Full article