Search Results (65)

In recent years, the proliferation of forged videos, known as deepfakes, has escalated significantly, primarily due to advancements in technologies such as Generative Adversarial Networks (GANs), diffusion models, and Vision Language Models (VLMs). These deepfakes present substantial risks, threatening political stability, facilitating celebrity impersonation, and enabling tampering with evidence. As the sophistication of deepfake technology increases, detecting these manipulated videos becomes increasingly challenging. Most of the existing deepfake detection methods use Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), or Vision Transformers (ViTs), achieving strong accuracy but exhibiting high computational demands. This highlights the need for a lightweight yet effective pipeline for real-time and resource-limited scenarios. This study introduces a lightweight deep learning model for deepfake detection in order to address this emerging threat. The model incorporates three integral components: MobileNet for feature extraction, a Convolutional Block Attention Module (CBAM) for feature enhancement, and a Gated Recurrent Unit (GRU) for temporal analysis. Additionally, a pre-trained Multi-Task Cascaded Convolutional Network (MTCNN) is utilized for face detection and cropping. The model is evaluated using the Deepfake Detection Challenge (DFDC) and Celeb-DF v2 datasets, demonstrating impressive performance, with 98.2% accuracy and a 99.0% F1-score on Celeb-DF v2 and 95.0% accuracy and a 97.2% F1-score on DFDC, achieving a commendable balance between simplicity and effectiveness. Full article

Network anomaly detection aims to identify abnormal traffic patterns that may indicate faults or cyber threats. This task requires modeling complex network flows composed of heterogeneous features, such as static headers, packet sequences, and statistical summaries. However, most existing methods focus on temporal modeling and treat flows as uniform sequences, overlooking feature heterogeneity and dependencies across domains. As a result, they often miss subtle anomalies that can be reflected by cross-domain correlations, highlighting the need for more structured modeling. We propose a domain-aware framework for network anomaly detection that explicitly models the heterogeneity of flow-level features and their cross-domain interactions. To address the limitations of prior work in handling heterogeneous flow features, we design an Intra-Domain Expert Network (IDEN) that uses convolutional and feed-forward layers to independently extract patterns from distinct domains. We further introduce an Inter-Domain Expert Network (EDEN) that uses attention mechanisms to capture dependencies across domains and produces integrated flow representations. These refined representations are passed to a Transformer-based temporal module to detect anomalies over time, including gradually evolving or coordinated behaviors. Experiments on multiple public datasets show that our method achieves higher detection accuracy, demonstrating the value of explicitly modeling intra-domain structure and inter-domain dependencies. Full article

Multidrug-resistant (MDR) Shiga toxin-producing Escherichia coli O157 (STEC O157) is a public health threat. This study analyzed publicly available surveillance data collected by the National Antimicrobial Resistance Monitoring System (NARMS) to assess temporal and regional differences and co-resistance patterns in MDR STEC O157 human clinical isolates across the United States. Co-resistance patterns were assessed by hierarchical clustering and Phi coefficient network analyses. A negative binomial regression model estimated the incidence rate ratios (IRRs) for the number of antimicrobial classes to which an isolate was resistant, across years and geographic regions. Out of 1955 isolates, 151 (7.57%) were MDR. The most important clusters were Cluster 1 (n = 1632), which included susceptible isolates, and Cluster 3 (n = 255), comprising the majority of the MDR isolates, having a high resistance prevalence to tetracyclines (TET) (0.97), folate pathway inhibitors (FPI) (0.77), and phenicols (PHN) (0.49). In the co-resistance network, TET, FPI, and PHN served as central hubs, with large nodes and thick edges, suggesting that they are frequently co-selected. The highest IRRs were observed in Regions 6 (IRR = 2.72) and 9 (IRR = 2.00), compared to Region 4. Compared to 2010, a significant increase in the IRR was observed in each year from 2015 to 2021 (IRRs 2.5–4.38). Antimicrobial stewardship programs and public health interventions targeting MDR E. coli O157 are needed to mitigate the emergence of antimicrobial resistance. Full article

Near-surface ozone (O₃) pollution has emerged as a growing threat to rice production in the Pearl River Delta (PRD), impairing photosynthesis, suppressing crop growth, and reducing yields. This study integrated long-term observational data with spatial crop distribution data and modeling approaches to assess O₃-induced impacts on rice yields and associated economic losses across the PRD from 2015 to 2023. The results showed that annual average O₃ concentrations during rice-growing periods increased from 41.3 to 66.0 μg/m³, with accumulated AOT40 values reaching 20.1 ppm·h. O₃ exposure led to annual average rice yield losses of 10.8% ± 0.8%, including 9.3% for double-early rice and 12.3% for double-late rice. Absolute yield losses totaled approximately 333,000 tons per year, equivalent to the caloric needs of 2.69 million people, with economic losses exceeding CNY 844 million. Vulnerability hotspots were identified in Zhaoqing and Jiangmen, each suffering over 100,000 tons of annual losses. Scenario simulations indicated that a 20% reduction in ambient O₃ could recover up to 54,700 tons annually. Future projections under RCP 2.6–8.5 suggested continued yield losses of 14,900 to 23,200 tons per year by 2050. Temporal adjustments to planting calendars may further mitigate these effects. This study highlights the urgent need for integrated mitigation strategies to enhance agricultural resilience in the face of ozone stress in industrialized delta regions. Full article

With the continuous growth of network threats, intrusion detection systems need to have robustness and adaptability to effectively identify malicious behaviors. However, factors such as noise interference, class imbalance, and complex attack pattern recognition have posed significant challenges to traditional systems. To address these issues, this paper proposes a dynamic preprocessing-enhanced DyP-CNX framework. The framework designs a sliding window dynamic interquartile range (IQR) standardization mechanism to effectively suppress the temporal non-stationarity interference of network traffic. It also combines a random undersampling strategy to mitigate the class imbalance problem. The model architecture adopts a CNN-XGBoost collaborative learning framework, combining a dual-channel convolutional neural network (CNN) and two-stage extreme gradient boosting (XGBoost) to integrate the original statistical features and deep semantic features. On the UNSW-NB15 and CSE-CIC-IDS2018 datasets, the method achieved F1 values of 91.57% and 99.34%, respectively. The experimental results show that the DyP-CNX method has the potential to handle the feature drift and pattern confusion problems in complex network environments, providing a new technical solution for adaptive intrusion detection systems. Full article

Assessing habitat quality and quantifying human disturbance are fundamental prerequisites for ecological conservation. However, existing studies predominantly focus on single dimensions. There is an urgent need to integrate habitat quality and human disturbance, and quantify their spatially coupled coordination relationships to address the disconnect between them in current research. As a critical ecological reserve in southeastern China, Fujian Province faces threats of ecological degradation. This study employed the InVEST model to evaluate habitat quality in Fujian from 1980 to 2020, utilized a human disturbance index to quantify spatiotemporal patterns of anthropogenic activities, analyzed their changes using landscape indices, and applied coupling coordination analysis to examine their interrelationships. Machine learning and geographically weighted regression were further integrated to identify driving factors of coupling coordination patterns. The results showed that: (1) Habitat quality remained relatively high while human disturbance levels stayed low throughout 1980–2020, though both showed gradual deterioration over time, particularly during 2010–2020, with riverine and coastal eastern regions exhibiting the lowest habitat quality and highest disturbance levels. (2) Coupling coordination relationships predominantly exhibited synergy, with moderate imbalance zones concentrated in the eastern coastal areas. Temporally, regions with lower imbalance expanded significantly during 2010–2020. (3) Landscape metric analysis revealed declining dominance of high-quality habitat/low-disturbance/synergistic zones, contrasted by increased fragmentation of low-quality habitat/high-disturbance/imbalanced zones. (4) Socioeconomic factors exerted stronger influence on coupling coordination patterns than natural environmental variables, proximity to urban areas, road density, and nighttime light indices. Each driver displayed spatially variable positive/negative effects. The results enhance our understanding of human–nature sustainable development dynamics, urban expansion–ecological conservation trade-offs, and provide methodological insights for regional ecological management and achieving sustainable development goals. Full article

Downhole torsional vibrations, especially high-frequency torsional oscillations (HFTOs) and stick–slip phenomena, pose a serious threat to drilling operations, often resulting in tool damage, prolonged non-productive time, and significant cost increases. Traditional monitoring methods cannot promptly capture complex vibration patterns, so there is an urgent need for advanced early warning systems. This study proposes the DS-DW-TimesNet model, which improves the TimesNet framework by incorporating downsampling technology for efficient data compression, dilated convolution that can expand the temporal receptive field, and a learnable weight normalization method that can stabilize the training process, thereby enhancing the capabilities of feature extraction and long-sequence modeling. Verified using field data from the Fuman Oilfield, the results show that in terms of the mean absolute error (MAE) for 210 s predictions, this model is 77.2% and 21.8% lower than LSTM and Informer, respectively, and the inference speed is increased by 78.5% (reaching 48 milliseconds). It can provide reliable 210 s early warning windows for high-frequency torsional oscillations and 150 s early warning windows for stick–slip, exceeding industry standards and helping to improve the safety and efficiency of drilling operations. Full article

Desertification poses critical threats to agricultural productivity and socio-economic stability, particularly in vulnerable regions like Thatta and Badin districts of Sindh, Pakistan. Traditional monitoring methods lack the accuracy and temporal resolution needed for effective early warning systems. This study presents a novel Spatio-Temporal Desertification Predictor (STDP) framework that integrates deep learning with next-generation satellite imaging for time-series desertification forecasting. The proposed encoder–decoder architecture combines Convolutional Neural Networks (CNNs) for spatial feature extraction from high-resolution satellite imagery with modified Long Short-Term Memory (LSTM) networks enhanced by multi-head attention to capture temporal dependencies. Environmental variables are fused through an adaptive data integration layer, and hyperparameter optimization is employed to enhance model performance for edge computing deployment. Experimental validation on a 15-year satellite dataset (2010–2024) demonstrates superior performance with MSE = 0.018, MAE = 0.079, and $R^2=0.94$, outperforming traditional CNN-only, LSTM-only, and hybrid baselines by 15–20% in prediction accuracy. The framework forecasts desertification trends through 2030, providing actionable signals for environmental management and policy-making. This work advances the integration of AI with satellite-based Earth observation, offering a scalable path for real-time environmental monitoring in IoT and edge computing infrastructures. Full article

Soil fugitive dust (SFD) emissions pose a significant threat to both human health and the environment, highlighting the need for accurate and reliable estimation and assessment in the desert regions of northwest China. This study used climate, soil, and vegetation data from Bayingolin Prefecture (2001–2022) and applied the WEQ model to analyze temporal and spatial variations in total suspended particulate (TSP), PM₁₀, and PM_2.5 emissions and their driving factors. The region exhibited high emission factors for TSP, PM₁₀, and PM_2.5, averaging 55.46 t km⁻² a⁻¹, 27.73 t km⁻² a⁻¹, and 4.14 t km⁻² a⁻¹, respectively, with pronounced spatial heterogeneity and the highest values observed in Yuli, Qiemo, and Ruoqiang. The annual average emissions of TSP, PM₁₀, and PM_2.5 were 3.23 × 10⁷ t, 1.61 × 10⁷ t, and 2.41 × 10⁶ t, respectively. Bare land was the dominant source, contributing 72.55% of TSP emissions. Both total emissions and emission factors showed an overall upward trend, reaching their lowest point around 2012, followed by significant increases in most counties during 2012–2022. Annual precipitation, wind speed, and temperature were identified as the primary climatic drivers of soil dust emissions across all counties, and their influences exhibited pronounced spatial heterogeneity in Bazhou. In Ruoqiang, Bohu, Korla, and Qiemo, dust emissions are mainly limited by precipitation, although dry conditions and sparse vegetation can amplify the role of wind. In Heshuo, Hejing, and Yanqi, stable vegetation helps to lessen wind’s impact. In Yuli, wind speed and temperature are the main drivers, whereas in Luntai, precipitation and temperature are both important constraints. These findings highlight the need to consider emission intensity, land use, or surface condition changes, and the potential benefits of increasing vegetation cover in severely desertified areas when formulating regional dust mitigation strategies. Full article

The sharp decline in the population along the northeastern border poses a significant threat to the security of the region, the prosperity of border areas, and the stability of the social economy in our country. Effective management of human and land resources is crucial for the high-quality development of border areas. Taking Changbai County on the northeastern border as an example, based on multi-source data such as land use, the natural environment, climate conditions, transportation location, and social economy from 2000 to 2020, the land use transfer matrix, spatial kernel density, and PLUS model were used to analyze the spatio-temporal evolution characteristics of land use and explore simulation scenarios and optimization strategies under different planning concepts. This study reveals the following: (1) During the study period, the construction land continued to increase, but the growth rate slowed down, mainly transferred from cultivated land and forest land, and the spatial structure evolved from a single center to a double center, with the core always concentrated along the border. (2) The distance to the port (transportation location), night light (social economy), slope (natural environment), and average annual temperature (climate conditions) are the main driving factors for the change in construction land, and the PLUS model can effectively simulate the land use trend under population contraction. (3) In the reduction scenario, the construction land decreased by 1.67 km², the scale of Changbai Town slightly reduced, and the contraction around Malugou Town and Badagou Town was more significant. The study shows that the reduction scenario is more conducive to the population aggregation and industrial carrying capacity improvement of shrinking county towns, which is in line with the high-quality development needs of border areas in our country. Full article

Frequent wildfires in the eastern grasslands of Mongolia pose significant threats to the ecological environment and pastoral livelihoods, creating an urgent need for high-temporal-resolution and high-precision fire prediction. To address this, this study established a daily-scale grassland fire risk assessment framework integrating multi-source remote sensing data to enhance predictive capabilities in eastern Mongolia. Utilizing fire point data from eastern Mongolia (2012–2022), we fused multiple feature variables and developed and optimized three models: random forest (RF), XGBoost, and deep neural network (DNN). Model performance was enhanced using Bayesian hyperparameter optimization via Optuna. Results indicate that the Bayesian-optimized XGBoost model achieved the best generalization performance, with an overall accuracy of 92.3%. Shapley additive explanations (SHAP) interpretability analysis revealed that daily-scale meteorological factors—daily average relative humidity, daily average wind speed, daily maximum temperature—and the normalized difference vegetation index (NDVI) were consistently among the top four contributing variables across all three models, identifying them as key drivers of fire occurrence. Spatiotemporal validation using historical fire data from 2023 demonstrated that fire points recorded on 8 April and 1 May 2023 fell within areas predicted to have “extremely high” fire risk probability on those respective days. Moreover, points A (117.36° E, 46.70° N) and B (116.34° E, 49.57° N) exhibited the highest number of days classified as “high” or “extremely high” risk during the April/May and September/October periods, consistent with actual fire occurrences. In summary, the integration of multi-source data fusion and Bayesian-optimized machine learning has enabled the first high-precision daily-scale wildfire risk prediction for the eastern Mongolian grasslands, thus providing a scientific foundation and decision-making support for wildfire prevention and control in the region. Full article

Air pollution poses a growing threat to public health and the environment, highlighting the need for continuous and precise urban air quality monitoring. The aim of this study was to implement and evaluate an autonomous air quality monitoring platform developed by the University of Ruse, “Angel Kanchev”, under Bulgaria’s National Recovery and Resilience Plan (project BG-RRP-2.013-0001), co-financed by the European Union through the NextGenerationEU initiative. The system, based on Libelium’s mobile sensor technology, was installed at a height of two meters on the university campus near Rodina Boulevard and operated continuously from 1 March 2024 to 30 March 2025. Every 15 min, it recorded concentrations of CO, CO₂, NO₂, SO₂, PM₁, PM_2.5, and PM₁₀, along with meteorological parameters (temperature, humidity, and pressure), transmitting the data via GSM to a cloud-based database. Analyses included a distributional assessment, Spearman rank correlations, Kruskal–Wallis tests with Dunn–Sidak post hoc comparisons, and k-means clustering to identify temporal and meteorological patterns in pollutant levels. The results indicate the high operational stability of the system and reveal characteristic pollution profiles associated with time of day, weather conditions, and seasonal variation. The findings confirm the value of combining calibrated IoT systems with advanced statistical methods to support data-driven air quality management and the development of predictive environmental models. Full article

The rapid expansion of Edge and Industrial Internet of Things (IIoT) systems has intensified the risk and complexity of cyberattacks. Detecting advanced intrusions in these heterogeneous and high-dimensional environments remains challenging. As the IIoT becomes integral to critical infrastructure, ensuring security is crucial to prevent disruptions and data breaches. Traditional IDS approaches often fall short against evolving threats, highlighting the need for intelligent and adaptive solutions. While deep learning (DL) offers strong capabilities for pattern recognition, single-model architectures often lack robustness. Thus, hybrid and optimized DL models are increasingly necessary to improve detection performance and address data imbalance and noise. In this study, we propose an optimized hybrid DL framework that combines a transformer, generative adversarial network (GAN), and autoencoder (AE) components, referred to as Transformer–GAN–AE, for robust intrusion detection in Edge and IIoT environments. To enhance the training and convergence of the GAN component, we integrate an improved chimp optimization algorithm (IChOA) for hyperparameter tuning and feature refinement. The proposed method is evaluated using three recent and comprehensive benchmark datasets, WUSTL-IIoT-2021, EdgeIIoTset, and TON_IoT, widely recognized as standard testbeds for IIoT intrusion detection research. Extensive experiments are conducted to assess the model’s performance compared to several state-of-the-art techniques, including standard GAN, convolutional neural network (CNN), deep belief network (DBN), time-series transformer (TST), bidirectional encoder representations from transformers (BERT), and extreme gradient boosting (XGBoost). Evaluation metrics include accuracy, recall, AUC, and run time. Results demonstrate that the proposed Transformer–GAN–AE framework outperforms all baseline methods, achieving a best accuracy of 98.92%, along with superior recall and AUC values. The integration of IChOA enhances GAN stability and accelerates training by optimizing hyperparameters. Together with the transformer for temporal feature extraction and the AE for denoising, the hybrid architecture effectively addresses complex, imbalanced intrusion data. The proposed optimized Transformer–GAN–AE model demonstrates high accuracy and robustness, offering a scalable solution for real-world Edge and IIoT intrusion detection. Full article

With the continuous development of industrial intelligence, the integration of cyber–physical components creates a need for effective attack detection methods to mitigate potential DDoS threats. Although several DDoS attack detection modeling approaches have been proposed, few effectively incorporate the unique characteristics of industrial control system (ICS) architectures and traffic patterns. This paper focuses on DDoS attack detection within cloud–edge collaborative ICSs and proposes a novel detection model called FedDynST. This model combines federated learning and deep learning to construct feature graphs of traffic data. Introducing dynamic and static adjacency matrices, this work reveals the interactions between long-term industrial traffic data and short-term anomalies associated with DDoS attacks. Convolutional neural networks are utilized to capture distinctive temporal features within industrial traffic, thereby improving the detection precision. Moreover, the model enables continuous optimization of the global detection framework through a federated learning-based distributed training and aggregation mechanism, ensuring the privacy and security of industrial client data. The effectiveness of the FedDynST model was validated on the CICDDoS2019 and Edge-IIoTset datasets. The simulation results validated the superiority of the proposed approach, and thus, demonstrated significant improvements in both detection accuracy and convergence. Full article

Abnormal ship navigation behaviors in cross-sea bridge waters pose significant threats to maritime safety, creating a critical need for accurate anomaly detection methods. Ship AIS trajectory data contain complex temporal features but often lack explicit labels. Most existing anomaly detection methods heavily rely on labeled or semi-supervised data, thus limiting their applicability in scenarios involving completely unlabeled ship trajectory data. Furthermore, these methods struggle to capture long-term temporal dependencies inherent in trajectory data. To address these limitations, this study proposes an unsupervised trajectory anomaly detection model combining a transformer architecture with a variational autoencoder (transformer–VAE). By training on large volumes of unlabeled normal trajectory data, the transformer–VAE employs a multi-head self-attention mechanism to model both local and global temporal relationships within the latent feature space. This approach significantly enhances the model’s ability to learn and reconstruct normal trajectory patterns, with reconstruction errors serving as the criterion for anomaly detection. Experimental results show that the transformer–VAE outperforms conventional VAE and LSTM–VAE in reconstruction accuracy and achieves better detection balance and robustness compared to LSTM–-VAE and transformer–GAN in anomaly detection. The model effectively identifies abnormal behaviors such as sudden changes in speed, heading, and trajectory deviation under fully unsupervised conditions. Preliminary experiments using the POT method validate the feasibility of dynamic thresholding, enhancing the model’s adaptability in complex maritime environments. Overall, the proposed approach enables early identification and proactive warning of potential risks, contributing to improved maritime traffic safety. Full article