Search Results (11,805)

This study investigates the integration of Circular Economy (CE) principles within the upstream beverage supply chain in Thailand, contextualized against a widening global circularity gap where macro rates have declined to 6.9% compared to 12.2% in the European Union. Moving beyond a general focus on underexplored stakeholders, this qualitative exploratory design examines the critical role of informal governance mechanisms in emerging markets. The research is grounded in a multi-theoretic framework integrating the Resource-Based View (RBV), Social Exchange Theory (SET), and Resource Dependence Theory (RDT). Thematic analysis was conducted on in-depth interviews with 23 key informants covering 18 core supply chain activities. The analytical results generated a three-tier hierarchical framework, culminating in a single overarching selective theme: Collaborative Upstream Resource Recirculation for Systemic Resilience. The findings reveal that circular supply chain performance is driven by the dynamic interplay between relational governance and internal resource capabilities, explicitly demonstrated by grassroots tactical innovations such as modifying production boilers to run on 100% biomass fuel. Relational trust and culturally embedded mechanisms (e.g., Sanya Jai) function as vital substitutes for formal institutional frameworks, enabling Supply Chain Collaboration (SCC) to drive adaptive practices and achieve system-level circular resilience. Full article

Imagery is fundamental to modern scientific research, making robust intrinsic camera calibration indispensable for accurate visual inference. The checkerboard-based calibration method has long been favored for its simplicity and ease of deployment and is widely used even in mission-critical computer vision pipelines. However, its limitations in modeling high-precision camera geometry can compromise downstream performance in tasks requiring geometric accuracy. In this work, camera calibration is revisited through the lens of photogrammetric self-calibration (PSC), and it is demonstrated that the PSC consistently outperforms the checkerboard method in both accuracy and precision across a range of vision tasks, including 3D reconstruction with structure from motion (SfM), visual simultaneous localization and mapping (SLAM), and novel-view synthesis and reconstruction. Our findings advocate for a paradigm shift toward calibration methods that better reflect the physical and projective properties of camera systems in real-world deployments for critical computer vision applications. Full article

Digital transformation is increasingly viewed as a strategic driver of operational efficiency, financial performance, and organisational resilience in rural transition economies. Existing research, however, largely assumes homogeneous digitalisation effects across firms while overlooking the structural conditions shaping integration efficiency. This study investigates the threshold relationship between supply chain integration and financial–economic performance using a threshold regression framework. The analysis is based on firm-level data from 80 agricultural, agritourism, and tourism-related firms operating in rural Northern Albania. Methodologically, the study combines Hansen’s threshold estimation with robust OLS and threshold logistic regression models, complemented by exploratory macro-level threshold analysis for Western Balkan economies. The findings reveal significant regime-dependent dynamics. Below the estimated socio-economic integration threshold, supply chain integration generates weak and statistically insignificant effects. Above the threshold, integration mechanisms produce substantially stronger financial and operational outcomes, indicating that digital transformation becomes economically productive primarily under sufficiently integrated organisational conditions. Additional diagnostics further show that highly integrated firms achieve superior coordination efficiency, resource allocation, and financial resilience. The study contributes to the literature by advancing a managerial-financial and coordination-based interpretation of digital transformation and its threshold performance effects in rural transition economies. Full article

Rare and safety-critical traffic situations remain challenging for autonomous driving (AD) because they are underrepresented in common training data and may include objects outside standard detector classes. This paper presents a real-time RGB-LiDAR fusion framework for detecting and reacting to rare traffic situations in CARLA (Car Learning to Act), a reproducible simulator for AD research. The approach combines YOLOv8n-based RGB perception, bird’s-eye-view (BEV) LiDAR clustering, decision-level fusion, an interpretable rule-based safety agent with hysteresis, Time-to-Collision (TTC)-aware escalation, and an automatic emergency braking (AEB) override above the CARLA autopilot. Fused observations are classified as semantic–geometric detections, semantic-only detections, or geometric-only obstacle candidates, where unmatched LiDAR clusters are treated conservatively as candidate-level physical evidence rather than confirmed rare objects. The framework was evaluated on three CARLA maps and 3CSim-inspired corner-case scenarios comprising 19,253 frames, with additional weather/lighting stress tests and a public nuScenes mini cross-platform check. On a manually annotated subset of 4800 CARLA frames, corresponding to approximately $24.9\%$ of the recorded CARLA log, the full framework achieved $96.2\%$ precision, $97.3\%$ recall, and a $96.7\%$ F1-score for safety-relevant threat detection. The control experiments show that the fusion-based safety agent reduced unnecessary braking to $1.7\%$ compared with $8.6\%$ for the LiDAR-only baseline and achieved event-level success on the annotated critical intervals. The proposed CPU-only implementation maintained real-time performance, with an average processing time of $34.7\mathrm{ms}$. Full article

The agricultural industry largely relies on conventional pesticides to maintain healthy, pest-free crops. Application of conventional insecticides is the go-to method for cultivating important food crops, such as corn and sorghum, free of Spodoptera frugiperda (fall armyworm) infestations. However, conventional insecticides have purported negative environmental and health impacts. Natural plant extracts, such as essential oils, are viewed as a promising alternative to conventional insecticides. In the current study, Dysphania ambrosioides (epazote) essential oil was embedded into an artificial diet and fed at two different concentrations to fall armyworms during a 10-day period. Final weights of the 5% epazote treatment group were statistically less (F₆₃₄₃ = 136.2 p < 0.001) than control groups. The 5% epazote treatment group also experienced the highest mortality rate (62%) of any treatment group (X² = 831.4, DF = 6, p < 0.001). Findings suggest that epazote essential oil has potential as an effective, natural insecticidal ingredient. This research is of importance to the fields of agronomy and health sciences. Full article

The internationalization of business-to-business (B2B) professional services is being reshaped by generative artificial intelligence (GenAI). Despite its potential to enhance productivity and reduce export uncertainty, existing research has focused on B2C contexts, leaving a gap in understanding how B2B experts perceive and exploit this technology. This research, using a Q methodology, seeks to explore the discursive framework of experts’ mindsets regarding exploitation of GenAI to develop B2B professional services exports. Using 32 experts from five countries (Iran, United States, United Kingdom, Germany, and India), four mindsets were identified: (1) Human–GenAI Synergy, (2) Export Innovation Catalyst, (3) Facilitator of Managerial Mindset, and (4) Moral Hazard Paradox. By conceptualizing mindsets as intangible resources within the Resource-Based View (RBV) and interpreting their role through the Uppsala model, this study makes three contributions: enriching theory through discourse-based analysis of expert mindsets, extending Q methodology to B2B export research, and providing practical insights for human–GenAI collaboration, export innovation, and ethical governance. The findings indicate that successful GenAI-enabled export development depends not only on technological capabilities but also on how experts interpret, adopt, and utilize the technology. The results highlight the need to balance innovation with ethical risks to achieve export growth. Full article

South Africa faces significant challenges in monitoring air pollution from different provinces due to the sparse nature of the sensor network and heterogeneous pollutant sources. Notably, some provinces continue to record a limited amount of data on air pollution, thus making monitoring in those locations problematic. Fortunately, the capabilities of deep learning models to facilitate effective monitoring in data-scarce locations have been highlighted by researchers; however, these models within the context of transfer learning still lack transparency and uncertainty quantification. Using air pollutants and meteorological factors, this study proposes a transfer learning model for particulate matter (PM_2.5) prediction in a data-scarce region. This transfer learning (TL) model leverages an adaptive Bi-directional Gated Recurrent Unit (adaBiGRU) with explainable artificial intelligence (xAI) and uncertainty quantification (UQ) to provide a novel uncertainty-aware adaptation transfer learning (UATL_adaBiGRU) model for a data-scarce location. Variant models based on the adaBiGRU technique, such as the temporal convolution network adaBiGRU (TCN-adaBiGRU) and domain-adversarial neural network adaBiGRU (DANNadaBiGRU), are presented as comparative models. The performance evaluation metrics are root mean squared, R² score and mean squared error. The R² score of pre-trained models in source domain is adaBiGRU (0.888), DANN_adaBiGRU (0.7788) and TCN_adaBiGRU (0.876). Furthermore, other comparative TL models include GRU (0.898), MLP (0.802) and adaptive LSTM (0.886). Afterwards, the pre-trained baseline model (adaBiGRU) was fine-tuned in the target domain dataset and the unpromising result contributed to the proposition of the UATL_adaBiGRU model for a data-scarce location, with R² score of 0.9618. Uncertainty assessment metrics results were also presented for the proposed model. Ablation assessment demonstrates that each component of the UATL_adaBiGRU contributes to enhancing the predictive performance. Again, the Diebold–Mariano (DM) test statistic demonstrates a statistically significant difference between baseline model and UATL_adaBiGRU model. Finally, the local interpretable model-agnostic explanation highlights multi-scaled features as contributing towards the prediction of PM_2.5 in the target domain. In view of this result, model fine-tuning is strongly recommended to enhance the robustness of the proposed uncertainty-aware adaption model in data-limited regions in South Africa. Full article

Accurate beef cattle body measurement data are crucial for growth assessment, phenotypic analysis, breeding management, and precision livestock farming. Traditional manual measurements are labor-intensive, time-consuming, and likely to cause stress in animals, making it difficult to meet the demands of large-scale livestock farming. This paper employs a structured systematic literature review method, in accordance with the PRISMA 2020 guidelines, to summarize research progress in vision-based beef cattle body measurement. This paper focuses on reviewing technical approaches such as 2D image-based measurement, 3D measurement using RGB-D and LiDAR, and multi-view fusion. It analyzes key technologies including image segmentation, keypoint detection, point cloud processing, 3D reconstruction, and geometric calculations, and compares the advantages and disadvantages of different methods in terms of measurement accuracy, robustness, cost, and farm applicability. The results indicate that 2D image-based methods are low-cost and flexible to deploy but have limited expressiveness for 3D body measurement parameters; RGB-D and LiDAR methods can provide spatial information but are affected by point cloud noise, occlusion, equipment costs, and data processing complexity; multi-view fusion can improve the completeness of body surface information but places high demands on calibration, registration, and system integration. Current research still faces challenges such as a lack of public datasets, inconsistent annotation standards, uncertainty regarding ground truth, insufficient cross-ranch generalization validation, and limited practical applications. Future research should focus on developing standardized datasets, conducting cross-scenario validation, advancing multimodal perception, creating lightweight models, and applying edge computing to drive the evolution of visual body measurement toward continuous monitoring and intelligent decision-making. Full article

Fentanyl is an extremely potent and highly lipophilic synthetic opioid, whose toxicity is marked by significant respiratory depression and central nervous system impairment. Naloxone is the primary antidote for opioid overdose; however, there is an increasing interest in intravenous lipid emulsion (ILE) as a supplementary therapeutic approach for poisoning with lipophilic agents. This study aimed to assess the antidotal effect of ILE in cases of experimental fentanyl toxicity and to compare its effectiveness with that of naloxone, both when administered alone and in combination. The research was performed on male Wistar rats. Various parameters were monitored, including heart rate, respiratory rate, nociceptive response (Hot Plate test), motor coordination (Rota-rod test), and behavioral metrics (Open Field test). Fentanyl induced significant cardiorespiratory depression and analgesia. Naloxone successfully counteracted the respiratory and nociceptive effects. ILE demonstrated positive effects on specific cardiorespiratory parameters, especially in the initial recovery phase, although its influence on analgesia was less pronounced and occurred later. The combination of naloxone and ILE appeared to promote earlier cardiorespiratory enhancement compared to naloxone used alone; nevertheless, these variations must be viewed with caution due to the brief duration of fentanyl’s effects and the absence of evidence supporting an increased maximal therapeutic benefit. These findings endorse the potential role of ILE as an adjunctive, rather than a standalone, antidotal treatment for acute fentanyl poisoning. Full article

The rapid development of deep learning methods has significantly improved the effectiveness of object detection in Earth Observation (EO) imagery. However, standard metrics such as Mean Average Precision (mAP) do not fully reflect their utility in operational analyses. This paper proposes a multi-stage methodology for evaluating vehicle detection models, combining classical evaluation with functional analysis encompassing object counting, density estimation, and occupancy index. The research was conducted on high-resolution imagery (WorldView, Pleiades) and the xView dataset, evaluating five YOLO variants alongside transformer-based and two-stage detectors under three training strategies, including fine-tuning. The results show that models achieving high mAP values (up to 0.952) can simultaneously produce significant errors in object count estimation. Models trained exclusively on xView exhibit a substantial performance drop (mAP@0.50 ≈ 0.45) under domain shift conditions. The best results were obtained using a fusion-based approach combining YOLOv9 and YOLOv12, which reduced the mean relative error to 0.14 and the counting error to 13 objects, maintaining a low density error (0.0023). Functional validation across 20 parking areas confirmed the stability of the proposed approach. The findings confirm that functional analysis constitutes a critical complement to classical evaluation in remote sensing applications. Full article

Quality of life is a multidimensional concept that includes characteristics of the residence and the wider living environment. The aim of the paper is to examine whether age groups differ in satisfaction with selected spatial characteristics of the residence and the immediate living environment, and whether comparable differences occur in the perceived healthy living environment. The research considers age and perceived quality of life on three distinct levels: specific spatial characteristics of the residence (for example, the presence of a balcony, terrace, daylighting and window view) with selected environmental characteristics of the immediate surroundings (open view, view of green areas, accessibility of green areas) and a broader perception of a healthy living environment (general perception of health and well-being). The research is based on questionnaire results that included 473 participants. The data were analyzed using analysis of variance, Tukey–Kramer comparisons and correlation analysis. The results show statistically significant differences between age groups, especially in regard to satisfaction with the presence of a balcony, terrace or atrium, to daylighting and the quality of the window view in the residences, with the satisfaction being lowest in the younger and highest in the older age group. Interestingly, there were no statistically significant differences in the general perception of a healthy living environment. The correlation analysis further showed that satisfaction and age was associated with home ownership, dwelling type, residential location and housing-cost burden, whereas proximity to green areas was not linearly associated with age. The findings showed that specific age groups perceived the quality of the living environment more pronouncedly when linked with specific spatial characteristics of the dwelling rather than the broader perception of the living environment. Full article

This research project aims to explore children’s multilingualism and how children’s preschools build on the language resources which families and children bring to their preschools. Research has shown that children actively create their own knowledge and can express themselves regarding their views on language learning and that each child has a unique bilingual or multilingual experience. This qualitative research study contributes to in-depth research with children on their multilingual experiences as described by their parents, preschool teachers and principals. Data was collected from 2022 to 2025 in semi-structured interviews with six parents, their children’s preschool teachers and principals in three preschools in Iceland. The findings indicate that the parents value their children’s language repertoire and use diverse tools to support their children’s multilingualism. The teachers in the study are all interested in supporting the children’s multilingualism, although some of them claim that they lack knowledge, training, and support in implementing multilingual practices. The children are active agents in developing language policies and practices in their families. They contribute to their families’ language practices with their input, ideas, choices, and language preferences. Although Icelandic is the main language used in the preschools, the children, supported by their teachers and principals, have the opportunity and agency to develop their multilingualism. The research findings provide practitioners, policy makers and parents with suggestions as to how children can and will benefit from support in maintaining their heritage languages. More structured collaboration and policy guidance are likely to strengthen children’s multilingualism. Full article

Multi-view learning can effectively exploit the consistency and complementarity among multiple data sources and has become a major research direction in semi-supervised classification. However, the existing methods commonly suffer from several limitations, including the loss of view-specific information caused by premature feature fusion, interference from redundant inter-view noise, and the limited discriminative capability of consensus representation. These issues severely restrict classification performance under low-label settings. To address these limitations, this paper proposes Dual-branch Multi-view Learning with Dual-contrastive Information Bottleneck. The proposed framework constructs a decoupled dual-branch graph convolutional architecture to explicitly separate view-specific representations from cross-view consensus representation, thereby alleviating feature homogenization at the structural level. Furthermore, we design a dual-contrastive information bottleneck optimization mechanism, where the CLUB constraint minimizes redundant mutual information across views to suppress noise, while the InfoNCE constraint maximizes the mutual information between consensus representation and labels to enhance discriminative capability. Additionally, we employ an adaptive attention fusion module to dynamically integrate the dual-branch representations, further refining task-relevant features. The experiments conducted on nine public datasets demonstrate that the proposed method achieves favorable performance improvements over most of the selected comparison methods in semi-supervised classification tasks. Full article

Children of imprisoned parents, who are often described as ‘orphans of justice’, suffer a multitude of disadvantages when a parent is sent to prison. Whilst their experiences of loss, stigma, and social exclusion are well documented, one area that remains critically under-examined is how court reporting processes may further exacerbate these harms. This study explores the lived experience of children 11–17 yrs (n = 6) who had experienced parental imprisonment, and non-offending adults (parents, caregivers, and adult children of offenders/n = 6) in relation to their experiences of parental crime reported in the press. This study also integrates views from individual interviews conducted with journalists and press regulators (n = 5), as well as data from a content analysis of three regional and two national newspapers across a three-week period. Findings indicate that current court reporting practices can be, for some children and family members, a contributing factor to their difficulties. Participating children and family members assert that publishing partial home addresses and references to family relationships heightens their visibility in the community, which they say contributes to community backlash and negatively affects their physical and mental wellbeing. The content analysis (n = 186 custody related news reports) showed selective disclosure of offenders’ personal and family details. Interviewed journalists strongly defended the principle of open justice and felt legally unable to add the wider context families often wished to share. However, they expressed genuine sympathy for the children, and while resistant to new legal restrictions, were open to developing voluntary guidance to help reduce harm where possible. This study proposes an integrated framework to strengthen ethical journalism and better protect children impacted by parental imprisonment, calling for improved public information, trauma-informed education, participatory research and practitioner tools that centre children’s rights. It argues that open justice must be balanced with relational accountability, ensuring open justice does not come at the expense of children’s wellbeing. Full article

The radiometric stability of satellite thermal infrared (TIR) channels is an indispensable prerequisite for the accurate retrieval of land surface temperature (LST) and the generation of reliable climate data records. This study evaluates the on-orbit radiometric calibration stability of the Fengyun-3D (FY-3D)/MEdium Resolution Spectral Imager-II (MERSI-II) TIR channels (channels 24 and 25) over four years (2021–2024) via a rigorous cross-calibration framework against Aqua/Moderate Resolution Imaging Spectroradiometer (MODIS). By imposing stringent spectral, spatial, temporal, and angular constraints to ensure the high fidelity of collocated pixel pairs, the cross-calibration results demonstrate that FY-3D/MERSI-II exhibits exceptional radiometric stability. Absolute brightness temperature biases are typically less than 0.1 K, with root mean square errors (RMSEs) limited to 1.20 K over a range of diurnal and seasonal conditions, demonstrating no noticeable systematic degradation. Furthermore, the downstream impact of this calibration on LST retrieval was quantified using the adapted National Oceanic and Atmospheric Administration Joint Polar Satellite System Enterprise algorithm. Validated against independent ground-based longwave radiation measurements collected from the Heihe Watershed Allied Telemetry Experimental Research network (HiWATER) and the Surface Radiation Budget Network (SURFRAD), the retrieved LST yielded overall biases of 0 K and −0.37 K, respectively, with RMSEs below 2.5 K. Cross-calibration demonstrates a limited and context-dependent impact on daytime LST, while the nighttime LST accuracy can be marginally improved using seasonal calibration coefficients derived from combined day/night matchups. Mechanistically, the integration of a soil directional emissivity model into the retrieval algorithm effectively mitigates viewing-zenith-angle (VZA)-induced uncertainties, systematically reducing biases by 0.12–0.20 K and RMSEs by 0.04–0.06 K. These findings confirm that the on-orbit radiometric calibration of FY-3D/MERSI-II meets scientific quality requirements and provide practical guidance for optimizing LST retrieval. Full article