Search Results (6,635)

Automated surface defect detection in steel manufacturing is pivotal for ensuring product quality, yet it remains an open challenge owing to the extreme heterogeneity of defect morphologies—ranging from hairline cracks and microscopic pores to elongated scratches and shallow dents. Existing approaches, whether classical vision pipelines or recent deep-learning paradigms, struggle to simultaneously satisfy the stringent demands of industrial scenarios: high accuracy on sub-millimeter flaws, insensitivity to texture-rich backgrounds, and real-time throughput on resource-constrained hardware. Although contemporary detectors have narrowed the gap, they still exhibit pronounced sensitivity–robustness trade-offs, particularly in the presence of scale-varying defects and cluttered surfaces. To address these limitations, we introduce MBY (MBDNet-Attention-YOLO), a lightweight yet powerful framework that synergistically couples the MBDNet backbone with the YOLO detection head. Specifically, the backbone embeds three novel components: (1) HGStem, a hierarchical stem block that enriches low-level representations while suppressing redundant activations; (2) Dynamic Align Fusion (DAF), an adaptive cross-scale fusion mechanism that dynamically re-weights feature contributions according to defect saliency; and (3) C2f-DWR, a depth-wise residual variant that progressively expands receptive fields without incurring prohibitive computational costs. Building upon this enriched feature hierarchy, the neck employs our proposed MultiSEAM module—a cascaded squeeze-and-excitation attention mechanism operating at multiple granularities—to harmonize fine-grained and semantic cues, thereby amplifying weak defect signals against complex textures. Finally, we integrate the Inner-SIoU loss, which refines the geometric alignment between predicted and ground-truth boxes by jointly optimizing center distance, aspect ratio consistency, and IoU overlap, leading to faster convergence and tighter localization. Extensive experiments on two publicly available steel-defect benchmarks—NEU-DET and PVEL-AD—demonstrate the superiority of MBY. Without bells and whistles, our model achieves 85.8% mAP@0.5 on NEU-DET and 75.9% mAP@0.5 on PVEL-AD, surpassing the best-reported results by significant margins while maintaining real-time inference on an NVIDIA Jetson Xavier. Ablation studies corroborate the complementary roles of each component, underscoring MBY’s robustness across defect scales and surface conditions. These results suggest that MBY strikes an appealing balance between accuracy, efficiency, and deployability, offering a pragmatic solution for next-generation industrial quality-control systems. Full article

The carbon dioxide spillover effects and influencing factors of the “Belt and Road Initiative” (BRI) in African countries must be assessed to evaluate the effectiveness, promote low-carbon transmissions in African countries, and provide recommendations for achieving the 2030 Sustainable Development Goals. This novel study employs carbon dioxide emission intensity (CEI) and per capita carbon dioxide emissions (PCE) as dual indicators to evaluate the spatial spillover effects of 54 BRI African countries on their neighboring countries’ carbon emissions from 2007 to 2023. It identifies the key factors and mechanisms affecting these spillover effects using the spatial differences-in-differences (SDID) model. Results indicate that since the launch of the BRI, the CEI and PCE of BRI African countries have significantly increased, largely due to trade patterns and industrialization structures. Greater trade openness has further boosted local economic development, thereby increasing carbon dioxide’s spatial spillover. Government management and corruption control levels show some heterogeneity in the spillover effects, which may be attributed to long-standing issues of weak institutional enforcement in Africa. Overall, this study reveals the complex relationship between BRI African economic development and environmental outcomes, highlighting the importance of developing sustainable development strategies and establishing strong differentiated regulatory regimes to effectively address environmental challenges. Full article

The rapid advancement of underwater wireless communication technologies is critical to unlocking the full potential of marine resource exploration and environmental monitoring. This paper reviews recent progress in three primary modalities: underwater acoustic communication, radio frequency (RF) communication, and underwater optical wireless communication (UWOC), each designed to address specific challenges posed by complex underwater environments. Acoustic communication, while effective for long-range transmission, is constrained by ambient noise and high latency; recent innovations in noise reduction and data rate enhancement have notably improved its reliability. RF communication offers high-speed, short-range capabilities in shallow waters, but still faces challenges in hardware miniaturization and accurate channel modeling. UWOC has emerged as a promising solution, enabling multi-gigabit data rates over medium distances through advanced modulation techniques and turbulence mitigation. Additionally, bio-inspired approaches such as electric field communication provide energy-efficient and robust alternatives under turbid conditions. This paper further examines the practical integration of these technologies in underwater platforms, including autonomous underwater vehicles (AUVs), highlighting trade-offs between energy efficiency, system complexity, and communication performance. By synthesizing recent advancements, this review outlines the advantages and limitations of current underwater communication methods and their real-world applications, offering insights to guide the future development of underwater communication systems for robotic and vehicular platforms. Full article

This review outlines a comprehensive methodology for the research and development of heterogeneous catalytic technologies (R&D_HeCaTe). Emphasis is placed on the fundamental interactions between reactants, solvents, and heterogeneous catalysts—specifically the roles of catalytic centers and support materials (e.g., functional groups) in modulating activation energies and stabilizing catalytic functionality. Particular attention is given to catalyst deactivation mechanisms and potential regeneration strategies. The application of molecular modeling and chemical engineering analyses, including reaction kinetics, thermal effects, and mass and heat transport phenomena, is identified as essential for R&D_HeCaTe. Reactor configuration is discussed in relation to key physicochemical parameters such as molecular diffusivity, reaction exothermicity, operating temperature and pressure, and the phase and “aggressiveness” of the reaction system. Suitable reactor types—such as suspension reactors, fixed-bed reactors, and flow microreactors—are evaluated accordingly. Economic and environmental considerations are also addressed, with a focus on the complexity of reactions, selectivity versus conversion trade-offs, catalyst disposal, and separation challenges. To illustrate the breadth and applicability of the proposed framework, representative industrial processes are discussed, including ammonia synthesis, fluid catalytic cracking, methanol production, alkyl tert-butyl ethers, and aniline. Full article

Cell confluence and number are critical indicators for assessing cellular growth status, contributing to disease diagnosis and the development of targeted therapies. Accurate and efficient cell segmentation is essential for quantifying these indicators. However, current segmentation methodologies still encounter significant challenges in addressing multi-scale heterogeneity, poorly delineated boundaries under limited annotation, and the inherent trade-off between computational efficiency and segmentation accuracy. We propose an innovative network architecture. First, a preprocessing pipeline combining contrast-limited adaptive histogram equalization (CLAHE) and Gaussian blur is introduced to balance noise suppression and local contrast enhancement. Second, a bidirectional feature pyramid network (BiFPN) is incorporated, leveraging cross-scale feature calibration to enhance multi-scale cell recognition. Third, adaptive kernel convolution (AKConv) is developed to capture the heterogeneous spatial distribution of glioma stem cells (GSCs) through dynamic kernel deformation, improving boundary segmentation while reducing model complexity. Finally, a probability density-guided non-maximum suppression (Soft-NMS) algorithm is proposed to alleviate cell under-detection. Experimental results demonstrate that the model achieves 95.7% mAP50 (box) and 95% mAP50 (mask) on the GSCs dataset with an inference speed of 38 frames per second. Moreover, it simultaneously supports dual-modality output for cell confluence assessment and precise counting, providing a reliable automated tool for tumor microenvironment research. Full article

This study proposes a novel Deep Reinforcement Learning (DRL) policy network structure for portfolio management called PortRSMs. PortRSMs employs stacked State-Space Models (SSMs) for the modeling of multi-scale continuous regime shifts in financial time series, striking a balance between exploring consistent distribution properties over short periods and maintaining sensitivity to sudden shocks in price sequences. PortRSMs also performs cross-asset regime fusion through hypergraph attention mechanisms, providing a more comprehensive state space for describing changes in asset correlations and co-integration. Experiments conducted on two different trading frequencies in the stock markets of the United States and Hong Kong show the superiority of PortRSMs compared to other approaches in terms of profitability, risk–return balancing, robustness, and the ability to handle sudden market shocks. Specifically, PortRSMs achieves up to a 0.03 improvement in the annual Sharpe ratio in the U.S. market, and up to a 0.12 improvement for the Hong Kong market compared to baseline methods. Full article

An in-depth understanding of the spatiotemporal heterogeneity of ecosystem service (ES) trade-offs and synergies, along with their driving factors, is crucial for formulating key ecological restoration strategies and effectively allocating ecological environmental resources in the Hulunbuir region. This study employed an integrated analytical approach combining the InVEST model, ArcGIS geospatial processing, R software environment, and Optimal Parameter Geographical Detector (OPGD). The spatiotemporal patterns and driving factors of the interaction of four major ES functions in Hulunbuir area from 2000 to 2020 were studied. The research findings are as follows: (1) carbon storage (CS) and soil conservation (SC) services in the Hulunbuir region mainly show a distribution pattern of high values in the central and northeast areas, with low values in the west and southeast. Water yield (WY) exhibits a distribution pattern characterized by high values in the central–western transition zone and southeast and low values in the west. For forage supply (FS), the overall pattern is higher in the west and lower in the east. (2) The trade-off relationships between CS and WY, CS and SC, and SC and WY are primarily concentrated in the western part of Hulunbuir, while the synergistic relationships are mainly observed in the central and eastern regions. In contrast, the trade-off relationships between CS and FS, as well as FS and WY, are predominantly located in the central and eastern parts of Hulunbuir, with the intensity of these trade-offs steadily increasing. The trade-off relationship between SC and FS is almost widespread throughout HulunBuir. (3) Fractional vegetation cover, mean annual precipitation, and land use type were the primary drivers affecting ESs. Among these factors, fractional vegetation cover demonstrates the highest explanatory power, with a q-value between 0.6 and 0.9. The slope and population density exhibit relatively weak explanatory power, with q-values ranging from 0.001 to 0.2. (4) The interactions between factors have a greater impact on the inter-relationships of ESs in the Hulunbuir region than individual factors alone. The research findings have facilitated the optimization and sustainable development of regional ES, providing a foundation for ecological conservation and restoration in Hulunbuir. Full article

In this paper a new state model is introduced, an adaptative state model in a binary temporal representation (ASMBRT) as well as its application in constructing an algorithmic trading system. The presented model uses the binary temporal representation, which allows for a precise analysis of exchange rates without losing any informative value of the data. The basis of the model is the trajectory analysis for the ensuing changes in price quotations and dependencies between the duration of each change. The main advantage of the model is to eliminate the threshold analysis, used in existing state models. This solution allows for a more accurate identification of investor behavior patterns, which translates into a reduction of investment risk. In order to verify obtained results in practice, the paper presents a concept of creating an algorithmic trading system and an analysis of its financial effectiveness for the exchange rate most popular among investors, namely EUR/USD. Full article

Target detection is essential in intelligent transportation and autonomous control of unmanned aerial vehicles (UAVs), with single-stage detection algorithms used widely due to their speed. However, these algorithms face limitations in detecting small targets, especially in aerial photography from unmanned aerial vehicles (UAVs), where small targets are often occluded, multi-scale semantic information is easily lost, and there is a trade-off between real-time processing and computational resources. Existing algorithms struggle to effectively extract multi-dimensional features and deep semantic information from images and to balance detection accuracy with model complexity. To address these limitations, we developed FFG-YOLO, a lightweight small-target detection method for UAVs based on YOLOv8. FFG-YOLO incorporates three modules: a feature enhancement block (FEB), a feature concat block (FCB), and a global context awareness block (GCAB). These modules strengthen feature extraction from small targets, resolve semantic bias in multi-scale feature fusion, and help differentiate small targets from complex backgrounds. We also improved the positioning accuracy of small targets using the Wasserstein distance loss function. Experiments showed that FFG-YOLO outperformed other algorithms, including YOLOv8n, in small-target detection due to its lightweight nature, meeting the stringent real-time performance and deployment requirements of UAVs. Full article

This paper investigates the cultural, spiritual, and ecological use and value of fish by-products in the material practices of Alaska Native (Indigenous Peoples are the descendants of the populations who inhabited a geographical region at the time of colonisation and who retain some or all of their own social, economic, cultural, and political institutions. In this paper, I use the terms “Indigenous” and “Native” interchangeably. In some countries, one of these terms may be favoured over the other.) and Greenlandic Inuit women. It aims to uncover how fish remnants—skins, bones, bladders, vertebrae, and otoliths—were transformed through tanning, dyeing, and sewing into garments, containers, tools, oils, glues, and adornments, reflecting sustainable systems of knowledge production rooted in Arctic Indigenous lifeways. Drawing on interdisciplinary methods combining Indigenist research, ethnographic records, and sustainability studies, the research contextualises these practices within broader environmental, spiritual, and social frameworks. The findings demonstrate that fish-based technologies were not merely utilitarian but also carried symbolic meanings, linking wearers to ancestral spirits, animal kin, and the marine environment. These traditions persisted even after European contact and the introduction of glass trade beads, reflecting continuity and cultural adaptability. The paper contributes to academic discourse on Indigenous innovation and environmental humanities by offering a culturally grounded model of zero-waste practice and reciprocal ecology. It argues that such ancestral technologies are directly relevant to contemporary sustainability debates in fashion and material design. By documenting these underexamined histories, the study provides valuable insight into Indigenous resilience and offers a critical framework for integrating Indigenous knowledge systems into current sustainability practices. Full article

Quantum finance, as a key application scenario of quantum computing, showcases multiple significant advantages of quantum machine learning over traditional machine learning methods. This paper first aims to overcome the limitations of the hidden quantum Markov model (HQMM) in handling continuous data and proposes an innovative method to convert continuous data into discrete-time sequence data. Second, a hybrid quantum computing model is developed to forecast stock market trends. The model was used to predict 15 stock indices from the Shanghai and Shenzhen Stock Exchanges between June 2018 and June 2021. Experimental results demonstrate that the proposed quantum model outperforms classical algorithmic models in handling higher complexity, achieving improved efficiency, reduced computation time, and superior predictive performance. This validation of quantum advantage in financial forecasting enables the practical deployment of quantum-inspired prediction models by investors and institutions in trading environments. This quantum-enhanced model empowers investors to predict market regimes (bullish/bearish/range-bound) using real-time data, enabling dynamic portfolio adjustments, optimized risk controls, and data-driven allocation shifts. Full article

As large language models (LLMs) increasingly enter specialized domains, inference without external resources often leads to knowledge gaps, opaque reasoning, and hallucinations. To address these challenges in power equipment defect grading, we propose a zero-shot question-answering framework that requires no task-specific examples. Our system performs two-stage retrieval—first using a Sentence-BERT model fine-tuned on power equipment maintenance texts for coarse filtering, then combining TF-IDF and semantic re-ranking for fine-grained selection of the most relevant knowledge snippets. We embed both the user query and the retrieved evidence into a Chain-of-Thought (CoT) prompt, guiding the pre-trained LLM through multi-step reasoning with self-validation and without any model fine-tuning. Experimental results show that on a held-out test set of 218 inspection records, our method achieves a grading accuracy of 54.2%, which is 6.0 percentage points higher than the fine-tuned BERT baseline at 48.2%; an Explanation Coherence Score (ECS) of 4.2 compared to 3.1 for the baseline; a mean retrieval latency of 28.3 ms; and an average LLM inference time of 5.46 s. Ablation and sensitivity analyses demonstrate that a fine-stage retrieval pool size of k = 30 offers the optimal trade-off between accuracy and latency; human expert evaluation by six senior engineers yields average Usefulness and Trustworthiness scores of 4.1 and 4.3, respectively. Case studies across representative defect scenarios further highlight the system’s robust zero-shot performance. Full article

This paper examines the changing role of emissions trading systems (ETSs) within the macro-financial framework of energy markets, emphasizing price dynamics and systemic spillovers. Utilizing monthly data from seven ETS jurisdictions spanning January 2021 to December 2024 (N = 287 observations after log transformation and first differencing), which includes four auction-based markets (United States, Canada, United Kingdom, South Korea), two secondary markets (China, New Zealand), and a government-set fixed-price scheme (Germany), this research estimates a panel vector autoregression (PVAR) employing a Common Correlated Effects (CCE) model and augments it with machine learning analysis utilizing XGBoost and explainable AI methodologies. The PVAR-CEE reveals numerous unexpected findings related to carbon markets: ETS returns exhibit persistence with an autoregressive coefficient of −0.137 after a four-month lag, while increasing inflation results in rising ETS after the same period. Furthermore, ETSs generate spillover effects in the real economy, as elevated ETSs today forecast a 0.125-point reduction in unemployment one month later and a 0.0173 increase in inflation after two months. Impulse response analysis indicates that exogenous shocks, including Brent oil prices, policy uncertainty, and financial volatility, are swiftly assimilated by ETS pricing, with effects dissipating completely within three to eight months. XGBoost models ascertain that policy uncertainty and Brent oil prices are the most significant predictors of one-month-ahead ETSs, whereas ESG factors are relevant only beyond certain thresholds and in conditions of low policy uncertainty. These findings establish ETS markets as dynamic transmitters of macroeconomic signals, influencing energy management, labor changes, and sustainable finance under carbon pricing frameworks. Full article

The pepper weevil Anthonomus eugenii is a devastating pest native to Central America that can cause severe damage to over 35 pepper varieties. Global trade in peppers has significantly increased the risk of its spread and expansion. Moreover, future climate change may add more uncertainty to its distribution, resulting in considerable ecological and economic damage globally. Therefore, we employed an ensemble model combining Random Forests and CLIMEX to predict the potential global distribution of A. eugenii in historical and future climate scenarios. The results indicated that the maximum temperature of the warmest month is an important variable affecting global A. eugenii distribution. Under the historical climate scenario, the potential global distribution of A. eugenii is concentrated in the Midwestern and Southern United States, Central America, the La Plata Plain, parts of the Brazilian Plateau, the Mediterranean and Black Sea coasts, sub-Saharan Africa, Northern and Southern China, Southern India, Indochina Peninsula, and coastal area in Eastern Australia. Under future climate scenarios, suitable areas in the Northern Hemisphere, including North America, Europe, and China, are projected to expand toward higher latitudes. In China, the number of highly suitable areas is expected to increase significantly, mainly in the south and north. Contrastingly, suitable areas in Central America, northern South America, the Brazilian Plateau, India, and the Indochina Peninsula will become less suitable. The total land area suitable for A. eugenii under historical and future low- and high-emission climate scenarios accounted for 73.12, 66.82, and 75.97% of the global land area (except for Antarctica), respectively. The high-suitability areas identified by both models decreased by 19.05 and 35.02% under low- and high-emission scenarios, respectively. Building on these findings, we inferred the future expansion trends of A. eugenii globally. Furthermore, we provide early warning of A. eugenii invasion and a scientific basis for its spread and outbreak, facilitating the development of effective quarantine and control measures. Full article

This research presents the development of a novel object detection algorithm designed to identify humans in natural outdoor environments using minimal computational resources. The proposed system, SARGAS, combines a custom convolutional neural network (CNN) classifier with MOG2 background subtraction and partial affine transformations for camera stabilization. A secondary CNN refines detections and reduces false positives. Unlike conventional supervised models, SARGAS is trained in a partially unsupervised manner, learning to recognize feature patterns without requiring labeled data. The algorithm achieved a recall of 93%, demonstrating strong detection capability even under challenging conditions. However, the overall accuracy reached 65%, due to a higher rate of false positives—an expected trade-off when maximizing recall. This bias is intentional, as missing a human target in search and rescue applications carries a higher cost than producing additional false detections. While supervised models, such as YOLOv5, perform well on data resembling their training sets, they exhibit significant performance degradation on previously unseen footage. In contrast, SARGAS generalizes more effectively, making it a promising candidate for real-world deployment in environments where labeled training data is limited or unavailable. The results establish a solid foundation for further improvements and suggest that unsupervised CNN-based approaches hold strong potential in object detection tasks. Full article