Search Results (630)

Facial expressions transmit information about internal states, both during social interaction and in response to shared stimuli such as films. When individuals view the same content, synchrony in their expressions reflects shared information processing, and the degree to which their expressions correlate indicates how similarly their perceptual and affective systems are responding to the common input. This makes interindividual expression synchrony a potential marker of engagement and subjective experience. However, the acquisition and analysis of facial data pose both ethical and technical challenges to researchers. ‘Trace’ is a research media player implemented in PsychoPy’s online platform Pavlovia, which captures anonymised facial landmark coordinates through a webcam, without the ethical and technical constraints of capturing and storing video images of participants. Nonetheless, its usefulness is currently limited due to the lack of available preprocessing and analysis tools. This paper describes the functionality of TraceLAB, a MATLAB-based toolbox designed for the preprocessing of Trace data: specifically, the formatting, aligning, and filtering of data. In addition, TraceLAB implements some novel analysis techniques to allow researchers to quantify interindividual synchrony of expressions (through correlated component analysis) and head movements (through Surrogate Synchrony), which may be interpreted as measures of shared information processing. These techniques are demonstrated here on both simulated and real datasets. Full article

This study presents the development and numerical investigation of a full-section fog curtain dust suppression system installed in the return airway of a fully mechanized longwall mining face, designed to mitigate airborne dust emissions escaping from the return airway during coal extraction. To optimize nozzle selection, comparative experiments were conducted under varying water pressure conditions. A porous medium model was employed to represent the dust capture mesh, enabling a systematic analysis of the pressure drop and airflow resistance characteristics across a range of wind velocities; the model parameters—viscous resistance coefficient (D) and inertial resistance coefficient (C₂)—were calibrated accordingly. Subsequently, coupled computational fluid dynamics simulations of fog dispersion and airflow fields were performed using a validated full-scale geometric model of the fully mechanized mining face. The influence of mesh pore size—via its effect on droplet size distribution uniformity—on the spatial distribution and velocity profile of the airflow field was quantitatively evaluated. The results show that the optimal spray nozzle was the fan-shaped atomizing spray nozzle, with a selected water pressure of 0.6 MPa. The droplet concentration in the porous media section increased from 0.026 kg∙m⁻³ to 0.052 kg∙m⁻³, and the volume share increased from 51.5% to 74.5%. The concentration of the filtered droplet increased from 0.00067 kg∙m⁻³ to 0.0013 kg∙m⁻³, and the size of particles adsorbed by the porous media increased from 140 μm in the proportion of most particles to 0.0013 kg∙m⁻³. The proportion of most particles above 140 μm was reduced to a range of 0–80 μm, and the optimal pore size was selected to be 100 mesh. Dust measurements were conducted at different measuring points in the return airway of the 25212 comprehensive mining face in the Hongliulin North plate area. The overall dust removal rates at points A, B, and C reached 88.90%, 83.71%, and 84.85%, and the respiratory dust removal rates reached 81.24%, 79.39%, and 80.33%, respectively, indicating that dust removal is effective. Full article

Diffuse pollution from agricultural runoff, characterized by intermittent discharges of complex contaminant mixtures, including nutrients, pesticides, and heavy metals (HMs), poses a persistent threat to global water quality. Conventional “end-of-pipe” strategies often fail to address these decentralized, nonpoint sources. This review examines the evolution of Permeable Reactive Barriers (PRBs) from static, abiotic filters into modern Permeable Reactive Bio-Barriers (PRBBs), engineered as dynamic, fixed-bed biofilm reactors. A key advancement in PRBB efficacy is the exploitation of biofilm plasticity, particularly in response to coexistence with organic and inorganic pollutants. While heavy metals are traditionally viewed as inhibitors, this review synthesizes evidence showing that subinhibitory HM levels can act as structural and functional drivers. These metals induce the upregulation of Extracellular Polymeric Substances (EPSs), creating a “protective shield” that sequesters metals and confers functional resilience on the microbial consortia responsible for nutrient removal and pesticide biodegradation. The review analyzes contaminant removal mechanisms, highlighting the bio-chemo synergy between reactive media and biofilms, and proposes a classification framework based on target contaminants, media, and technological integration. Significant focus is placed on emerging hybrid multi-media systems designed to protect the microbial community from toxic metal shocks, alongside the integration of artificial intelligence for predictive control. While challenges in hydraulic sustainability and field validation remain, PRBBs represent a compact, low-energy, and scalable ecotechnology. PRBBs offer a strategically targeted solution within the Nature-Based Solutions toolkit for building resilient protection of aquatic ecosystems at the critical land-water interface. Full article

Microplastics (MPs) are emerging pollutants of global concern, posing significant ecological and human health risks. They are frequently detected in stormwater systems, with urban runoff serving as a major transport pathway into the environment. Green stormwater infrastructure, particularly bioretention systems (BRSs), offers a promising approach to mitigate these risks by filtering and retaining various contaminants. However, the occurrence of MPs in BRSs and their capacity to retain these pollutants remain largely unexplored in the literature, despite being critical for stormwater management and water quality protection. Therefore, this study attempted to examine the occurrence, vertical distribution, and trapping of MPs within a field-installed BRS, potentially emphasizing their role in reducing microplastic (MP) transport. Therefore, field samples were collected at depths of 2, 12, and 24 inches below the surface and processed in the laboratory for MP detection and quantification. The results revealed an average concentration of 1095 particles per kg of dried sediment, with fragments (microplastics shape) accounting for 78.54% of the total MPs. Although no clear vertical distribution pattern was observed, the initial findings showed that MPs were mostly retained at 24 inches, potentially indicating their transport through the media and the retention capacity of a BRS (surface and middle layer) in capturing microplastics from stormwater environments. However, there is no direct evidence to explain the mechanisms driving the observed concentrations at greater depths. The preliminary findings of this study highlight that the concentrations of different sizes of MPs can vary with soil depth in bioretention media. Integrating a BRS into urban stormwater infrastructure likely provides the dual benefits of improved stormwater management and reduced plastic pollution. This study underscores the importance of optimizing bioretention design and media composition to enhance MP trapping from stormwater. Full article

Multimode fibers with core diameters of 50 μm and 62.5 μm are the core media for short-distance, low-cost, and high-bandwidth optical transmission scenarios. Currently, the detection of their end-face defects is still mainly based on manual microscopic inspection. Most of the existing machine vision detection schemes are aimed at polarization-maintaining fibers (POL), which are easily interfered with by impurities and have insufficient accuracy and efficiency. This study introduces the information entropy in information theory as a constraint for feature selection, proposes the WGMOS digital image detection method, and optimizes the entire process of image acquisition, correction, filtering, adaptive segmentation, and feature extraction. By minimizing the information entropy of background noise and maximizing the information content of defect features, interference is suppressed. Experiments show that compared with the POL detection method, this scheme can exclude more impurities, with the image equalization value increased by ≥38.20% and the signal-to-noise ratio increased by ≥6.0%. It can achieve efficient and accurate detection of multimode fiber end-face defects. Full article

This paper presents an edge-deployable vision-based framework for human–robot interaction using a xArm collaborative robot and a single RGB camera mounted on the robot wrist, and lightweight AI-based perception modules. The system enables intuitive, contact-free control by combining hand understanding and object detection within a unified perception–decision–control pipeline. Hand landmarks are extracted using MediaPipe Hands, from which continuous hand trajectories, static gestures, and dynamic gestures are derived. Task objects are detected using a YOLO-based model, and both hand and object observations are mapped into the robot workspace using ArUco-based planar calibration. To ensure stable robot motion, the hand control signal is smoothed using low-pass and Kalman filtering, while dynamic gestures such as waving are recognized using a lightweight LSTM classifier. The complete pipeline runs locally on edge hardware, specifically NVIDIA Jetson Orin Nano and Raspberry Pi 5 with a Hailo AI accelerator. Experimental evaluation includes trajectory stability, gesture recognition reliability, and runtime performance on both platforms. Results show that filtering significantly reduces hand-tracking jitter, gesture recognition provides stable command states for control, and both edge devices support real-time operation, with Jetson achieving consistently lower runtime than Raspberry Pi. The proposed system demonstrates the feasibility of low-cost edge AI solutions for responsive and practical human–robot interaction in collaborative industrial environments. Full article

Deepfake technology poses significant threats to information authenticity and social trust, necessitating effective detection methods. However, existing detection approaches predominantly rely on high-complexity network architectures that, while accurate in controlled environments, suffer from prohibitive computational costs that hinder deployment in resource-constrained scenarios such as social media platforms. To address this efficiency-accuracy dilemma, we propose a lightweight face forgery detection method that systematically learns multi-scale forgery traces. Our approach features a four-stage lightweight architecture that hierarchically extracts features from local textures to global semantics, mimicking the human visual system. Within each stage, a multi-scale dynamic perception mechanism divides feature channels into parallel groups equipped with lightweight attention modules to capture forgery cues spanning pixel-level anomalies, local structures, regional patterns, and semantic inconsistencies. Furthermore, rather than relying on conventional feature fusion that risks suppressing subtle artifacts, we introduce a novel Context-Guided Dynamic Convolution. This mechanism uses mid-level spatial anomalies as active anchors to dynamically modulate high-level semantic filters, with the goal of mitigating the disconnect between semantic content and forgery evidence. Our model achieves strong performance, yielding an AUC of 91.98% on FaceForensics++ and 93.50% on DeepFake Detection Challenge, outperforming current state-of-the-art lightweight methods. Furthermore, compared to heavy Vision Transformers, our model achieves a superior performance-efficiency trade-off, requiring only 3.06 M parameters and 1.36 G FLOPs, making it highly suitable for real-time, resource-constrained deployment. Full article

Background/Objectives: Gastro-resistant multiparticulate systems are designed to protect drugs in acidic environments and to ensure intestinal release. In practice, the method of administration may need to be modified: pellet-containing capsules opened or tablets halved for patients with swallowing difficulties, yet the type of liquid used for administration is often not specified. This study examined the stability of gastro-resistant coated pellets after exposure to various aqueous media prior to ingestion. Methods: To evaluate administration instructions, 103 Summaries of Product Characteristics of gastro-resistant products were reviewed. Pellets were produced using a bottom-spray fluidized bed process and coated with Eudragit L 30 D-55. Dissolution testing in pH 1.2 medium was performed after pre-soaking the pellets for 5, 15, and 30 min in beverages with various pH and conductivity. Drug release was measured by UV-VIS method, and morphological changes were assessed by image analysis. Marketed gastro-resistant products were also examined visually. Results: SmPC review revealed that the beverage for intake was frequently unspecified. Among the tested beverages differences in pH and conductivity were observed. Alkaline medicinal mineral waters induced increased and time-dependent premature drug release compared to tap and filtered water. Image analysis indicated a reduction in surface area after exposure to alkaline media. Conclusions: Contact with non-specified aqueous media before swallowing may weaken the protective function of gastro-resistant films. More explicit recommendations on suitable administration manipulation and media may improve therapeutic consistency. Full article

Public opinion on online platforms now plays an important role in agricultural markets, which have always been unpredictable. Although sentiment analysis has been widely applied to agricultural texts, most existing studies typically focus only on classification accuracy without connecting results to actual market intelligence systems, especially in multilingual contexts. This paper introduces a reliability-aware transformer-based framework for analyzing sentiment in agricultural market intelligence across multiple languages. The framework leverages weakly supervised multilingual transformers to extract sentiment signals from large-scale unlabeled Thai and English texts about major agricultural commodities found online. To enhance robustness under weak supervision, the framework incorporates reliability-aware mechanisms, including confidence-based pseudo-label filtering, cross-source consistency refinement, and expert-guided calibration to reduce noise and account for bias between different data sources. Sentiment predictions are further aligned with market intelligence objectives through reliability-weighted aggregation, yielding interpretable sentiment indices that enable cross-lingual and cross-source comparability. We tested the framework extensively using a multilingual agricultural corpus derived from social media and news coverage of agriculture. The results show consistent improvements over both classical machine learning approaches and standard multilingual transformer baselines. Additional ablation studies and sensitivity analyses confirmed that reliability-aware mechanisms, particularly confidence thresholding, play a crucial role in getting the right balance between label quality and data coverage. Overall, the results indicate that reliability-aware multilingual sentiment analytics provide robust and actionable insights for agricultural market monitoring and policy analysis. Full article

High-temperature particulate matter (PM) filtration remains a fundamental challenge, because most fiber filters not only face the challenge of high temperatures but also suffer from an inherent trade-off between capture efficiency, pressure drop, and service life. This paper reports a hierarchical layered zirconia (ZrO₂) ceramic fiber aerogel featuring a continuous multiscale gradient. The aerogel was prepared by gradient air-blown spinning, and the resulting structure has directional order, with the fiber diameter gradually decreasing from upstream to downstream, thus forming a pore size gradient and achieving hierarchical particle interception across multiple scales. This rational design simultaneously suppresses surface clogging and reduces flow resistance, resolving the longstanding trade-off between efficiency and permeability. Consequently, this aerogel achieves an ultra-high filtration efficiency of 99.96%, a low pressure drop of 156 Pa, and a high dust-holding capacity of 101 g m⁻². The material also exhibits outstanding mechanical toughness (80% compressive strain elasticity and 25.75% tensile fracture strain) and thermal stability up to 1000 °C. Moreover, it maintains over 99.95% filtration efficiency at high temperatures and can be fully regenerated through 800 °C heat treatment. This work establishes a structure-based design paradigm for high-temperature filtration media and provides a scalable pathway for next-generation industrial flue gas purification. Full article

Quorum quenching (QQ) is a promising biological approach that has the potential to control membrane biofouling. However, the implementation of the QQ membrane bioreactor still requires a more systematic and comprehensive understanding, including the selection of membrane materials, the determination of the optimal QQ bacterial dosage, and the use of appropriate media for the immobilization of QQ bacteria, all of which are important to ensure long-term operation. The present study investigated the impact of QQ bacteria on biofilm formation across different polymeric membranes. These include flat sheet membranes, Polytetrafluoroethylene (PTFE), Polysulfones (PSs), and hollow-fibre polyvinylidene difluoride (PVDF) membranes. It also evaluated biofilm development, membrane filtration performance, extracellular polymeric substance (EPS) production, and sludge floc properties, which were characterized using fluorescence microscopy. The results revealed that QQ intervention markedly suppressed quorum sensing (QS), leading to a pronounced, dose-dependent reduction in biofilm thickness, membrane fouling, EPS production and sludge floc size. Biofilm thickness was reduced by 63.5% on PTFE and 55.4% on PS membranes, accompanied by a notable reduction in EPS protein and polysaccharides, thereby weakening the biofilm formation and enhancing membrane filterability. Therefore, the permeability performance of the PVDF membrane improved by 338.2%. Furthermore, sludge settleability was enhanced, and floc size was reduced, resulting in the mitigation of biofilm formation without impacting pollutant degradation. These findings elucidate the material-dependent and dose-responsive mechanism by which QQ regulates EPS synthesis and biofilm formation in MBR. Full article

Phosphorus is a ubiquitous contaminant in urban and agricultural landscapes. A retention basin located in the southern part of Barnegat Bay, New Jersey, was identified as receiving stormwater runoff with elevated phosphorus concentrations. The basin is surrounded by expanding urban development, contributing to the progressive degradation of water quality in the bay, which is already highly eutrophic. This study evaluated the effectiveness of a culvert retrofit with a green filter media composed of granulated-aluminum-based drinking water-treatment residuals (Al-WTR) and granular carbon (5:1 ratio, w/w) for the removal of phosphorus and suspended sediments from stormwater runoff. The performance of the filter media was assessed through water quality monitoring following runoff events over a 12-month period. The results indicated that the green filter media achieved up to 52% removal of total phosphorus from stormwater influent. However, treatment efficiency declined after approximately five months due to clogging of the geotextile bag housing the media. The replacement of the geotextile bag restored phosphorus removal performance (59%), highlighting the importance of routine maintenance. The findings demonstrate a cost-effective, environmentally sustainable, and innovative green engineering approach for mitigating phosphorus contamination in urban stormwater. Full article

The purpose of this paper is to analyze the media visibility of investments implemented in Wrocław, with a particular focus on the democratization of urban processes through the Wrocław Participatory Budget (WPB) and to study the public perception of these projects within the local information landscape. The paper presents an integrated analytical methodology combining geospatial data from the Spatial Information System of Wrocław (SIP) with textual data from the full corpus of local news articles from Wrocław. A hybrid data processing pipeline was used, including filtering of articles about Wrocław, geoparsing of location names, matching articles to investments using classic Term Frequency-Inverse Document Frequency (TF-IDF) models and embedding in language models such as HerBERT, and sentiment analysis using the XLM-T model. The results reveal strong imbalances in the visibility of WPB projects, that almost 90% of investments were not mentioned even once in the media. Temporal sentiment analysis indicated differences between categories of WPB projects. The results confirm the existence of “media deserts” and “islands of attention,” which leads to information exclusion for specific local communities and marginalized groups. This translates into asymmetry in residents’ knowledge of the real scope of the WPB program. The paper emphasizes the importance of Geographic Information System (GIS) fusion methods with natural language processing models (NLP) for urban research, and identifies directions for further analysis, including accompanying problems and limitations in the present day. Full article

Millions of people across rural and peri-urban regions worldwide remain exposed to unsafe concentrations of naturally occurring fluoride in groundwater. In West Bengal, India, community-level water purification plants (CWPPs) have been widely installed to remove excess fluoride, yet their long-term operational performance remains minimally documented. This study assessed the pre-filter and post-filter water quality of 58 such groundwater-based CWPPs across the fluoride-affected districts of Bankura and Purulia in West Bengal, to evaluate in-field fluoride removal performance and potential hydrogeochemical, operational, and management drivers. Evaluation included fluoride concentration and key physicochemical parameters such as pH, temperature, electrical conductivity (EC), oxidation-reduction potential (ORP), total dissolved solids (TDS), and other anions including bromide, chloride, bicarbonate, nitrite, nitrate, phosphate, and sulphate. Fluoride concentration ranged from 1.7 mg/L to 8.2 mg/L and 1.6 mg/L to 3.9 mg/L in the sampled source water of Bankura and Purulia respectively, with both pre- and post-filter water of all the observed treatment units exceeding the WHO guideline of 1.5 mg/L. Potential contributors to underperformance may include inappropriate filter media selection, insufficient backwashing and regeneration, limited operational oversight and/or non-tailored treatment approaches. However, details on the adsorbent media and operational details were not available, and thus findings reflect observed field performance rather than necessarily causal relationships. These operational insights will contribute to the global discussion on improving decentralized groundwater treatment systems in resource-constrained settings. Full article

Rumor source detection aims to identify the initial origin of misinformation diffusion in social networks. Accurate source localization is essential for effective rumor intervention and early mitigation in large-scale social media platforms. Existing rumor source detection methods often struggle to model complex propagation structures. However, applying mathematical models uniformly to all samples introduces unnecessary computational overhead and limits scalability. By leveraging GNN-based candidate ranking, our approach effectively narrows the source search space and provides a reliable structural foundation for subsequent reasoning. Prior studies typically perform end-to-end inference without considering prediction confidence, leading to inefficient processing of low-uncertainty samples. To address this issue, we introduce an entropy-based uncertainty filtering mechanism that selectively identifies high-uncertainty cases requiring further reasoning, significantly reducing redundant computation. Meanwhile, existing methods lack semantic interpretability when handling ambiguous propagation patterns, motivating the incorporation of large language model (LLM) reasoning. We employ LLM-based reasoning only on filtered samples to enhance semantic understanding while controlling inference cost. Based on these designs, we propose TSR-RSD, a tri-stage selective reasoning framework that integrates GNN-based structural modeling, uncertainty-driven sample selection, and LLM-based semantic reasoning. Experimental results on GossipCop, PolitiFact, and PHEME demonstrate that TSR-RSD consistently outperforms GNN-based baselines in terms of Hit@1, Hit@3, Hit@5, and Mean Reciprocal Rank (MRR), reflecting improved accuracy and stability in rumor source ranking. Furthermore, the entropy-based uncertainty filtering mechanism significantly reduces the LLM invocation ratio by approximately 40–60%, while maintaining comparable or improved ranking performance. As a result, TSR-RSD achieves an overall inference time reduction of 35–50%, effectively balancing localization accuracy, computational efficiency, and interpretability. Full article