Search Results (1,632)

Forest surveying and inspection face significant challenges due to unstructured environments, variable terrain conditions, and the high costs of manual data collection. Although mobile robotics and artificial intelligence offer promising solutions, reliable autonomous navigation in forest, terrain-aware path planning, and tree parameter estimation remain open challenges. In this paper, we present the results of the AI4FOREST project, which addresses these issues through three main contributions. First, we develop an autonomous mobile robot, integrating SLAM-based navigation, 3D point cloud reconstruction, and a vision-based deep learning architecture to enable tree detection and diameter estimation. This system demonstrates the feasibility of generating a digital twin of forest while operating autonomously. Second, to overcome the limitations of classical navigation approaches in heterogeneous natural terrains, we introduce a machine learning-based surrogate model of wheel–soil interaction, trained on a large synthetic dataset derived from classical terramechanics. Compared to purely geometric planners, the proposed model enables realistic dynamics simulation and improves navigation robustness by accounting for terrain–vehicle interactions. Finally, we investigate the impact of point cloud density on the accuracy of forest parameter estimation, identifying the minimum sampling requirements needed to extract tree diameters and heights. This analysis provides support to balance sensor performance, robot speed, and operational costs. Overall, the AI4FOREST project advances the state of the art in autonomous forest monitoring by jointly addressing SLAM-based mapping, terrain-aware navigation, and tree parameter estimation. Full article

In recent decades, multiple approaches have emerged to assess architectural visual character, including fractal dimension analysis, visual equilibrium calculations, and visual preference surveys. However, the relationships among these methods and their alignment with subjective perception remain unclear. This study applies all three techniques to sample mosques in Riyadh, Saudi Arabia, to evaluate their validity and interconnections. Findings reveal a within-sample tendency toward low visual complexity, with fractal dimensions ranging from 1.2 to 1.547. Within this small, exploratory sample of five large main-road mosques in Riyadh, correlations between computed visual equilibrium and survey results provide preliminary, sample-specific convergent-validity evidence for Larrosa’s visual-forces method, rather than general validation. Within this sample, traditional façades with separate minarets tended to score as more visually balanced than more contemporary compositions. This triangulated approach offers an exploratory framework for architectural visual assessment that integrates objective metrics with human perception. Full article

Wild ginkgo, as an endangered species, holds significant value for genetic resource conservation, yet its practical applications face numerous challenges. Traditional field surveys are inefficient in mountainous mixed forests, while satellite remote sensing is limited by spatial resolution. Current deep learning approaches relying on single-source data or merely simple multi-source fusion fail to fully exploit information, leading to suboptimal recognition performance. This study presents a multimodal ginkgo crown dataset, comprising RGB and multispectral images acquired by an UAV platform. To achieve precise crown segmentation with this data, we propose a novel dual-branch dynamic weighting fusion network, termed dual-branch cross-modal attention-enhanced UNet (DCA-UNet). We design a dual-branch encoder (DBE) with a two-stream architecture for independent feature extraction from each modality. We further develop a cross-modal interaction fusion module (CIF), employing cross-modal attention and learnable dynamic weights to boost multi-source information fusion. Additionally, we introduce an attention-enhanced decoder (AED) that combines progressive upsampling with a hybrid channel-spatial attention mechanism, thereby effectively utilizing multi-scale features and enhancing boundary semantic consistency. Evaluation on the ginkgo dataset demonstrates that DCA-UNet achieves a segmentation performance of 93.42% IoU (Intersection over Union), 96.82% PA (Pixel Accuracy), 96.38% Precision, and 96.60% F1-score. These results outperform differential feature attention fusion network (DFAFNet) by 12.19%, 6.37%, 4.62%, and 6.95%, respectively, and surpasses the single-modality baselines (RGB or multispectral) in all metrics. Superior performance on cross-flight-altitude data further validates the model’s strong generalization capability and robustness in complex scenarios. These results demonstrate the superiority of DCA-UNet in UAV-based multimodal ginkgo crown recognition, offering a reliable and efficient solution for monitoring wild endangered tree species. Full article

Maritime transportation accounts for approximately 80 percent of global trade volume, with modern vessels increasingly reliant on Software-Defined Radio (SDR) technologies for communication and navigation. However, the very flexibility and reconfigurability that make SDRs advantageous also introduce complex radio frequency vulnerabilities exposing ships to threats that jeopardize vessel security, and this disrupts global supply chains. This survey paper systematically examines the security landscape of maritime SDR systems through a threat modeling lens. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we analyzed 84 peer-reviewed publications (from 2002 to 2025) and applied the STRIDE framework to identify and categorize maritime SDR threats. We identified 44 distinct threat types, with tampering attacks being most prevalent (36 instances), followed by Denial of Service (33 instances), Repudiation (30 instances), Spoofing (23 instances), Information Disclosure (24 instances), and Elevation of Privilege (28 instances). These threats exploit vulnerabilities across device, software, network, message, and user layers, targeting critical systems including Global Navigation Satellite Systems, Automatic Identification Systems, Very High Frequency or Digital Selective Calling systems, Electronic Chart Display and Information Systems, and National Marine Electronics Association 2000 networks. Our analysis reveals that maritime SDR threats are multidimensional and interdependent, with compromises at any layer potentially cascading through entire maritime operations. Significant gaps remain in authentication mechanisms for core protocols, supply chain assurance, regulatory frameworks, multi-layer security implementations, awareness training, and standardized forensic procedures. Further analysis highlights that securing maritime SDRs requires a proactive security engineering that integrates secured hardware architectural designs, cryptographic authentications, adaptive spectrum management, strengthened international regulations, awareness education, and standardized forensic procedures to ensure resilience and trustworthiness. Full article

Global Navigation Satellite Systems (GNSS) have become essential tools in geomatics engineering for precise positioning, cadastral surveys, topographic mapping, and deformation monitoring. Recent advances integrate GNSS with emerging technologies such as artificial intelligence (AI), machine learning (ML), cloud computing, and unmanned aerial systems (UAS), which have greatly improved accuracy, efficiency, and analytical capabilities in managing geospatial big data. In this study, we propose a hybrid Convolutional Neural Network–Long Short Term Memory (CNN-LSTM) architecture for seismic detection using high-rate (5 Hz) GNSS velocity time series. The model is trained on a large synthetic dataset generated by and real high-rate GNSS non-event data. Model performance was evaluated using real event and non-event data through an event-based approach. The results demonstrate that a hybrid deep-learning architecture can provide a reliable framework for seismic detection with high-rate GNSS velocity time series. Full article

Diplodia shoot blight is an opportunistic fungal pathogen infesting many conifer species and it has a global distribution. Depending on the duration and severity of the disease, affected needles appear yellow (chlorotic) for a brief period before becoming red or brown in colour. These symptoms can occur on individual branches or over the entire crown. Aerial sketch-mapping or the manual interpretation of aerial photography for tree health surveys are labour-intensive and subjective. Recently, however, the application of deep learning (DL) techniques to detect and classify tree crowns in high-spatial-resolution imagery has gained significant attention. This study evaluated two complementary DL approaches for the detection and classification of Pinus radiata trees infected with diplodia shoot blight across five geographically dispersed sites with varying topographies over two acquisition years: (1) object detection using YOLOv12 combined with Segment Anything Model (SAM) and (2) pixel-level semantic segmentation using U-Net, SegFormer, and EVitNet. The three damage classes for the object detection approach were ‘yellow’, ‘red-brown’ (both whole-crown discolouration) and ‘dead tops’ (partially discoloured crowns), while for the semantic segmentation the three classes were yellow, red-brown, and background. The YOLOv12m model achieved an overall mAP50 score of 0.766 and mAP50–95 of 0.447 across all three classes, with red-brown crowns demonstrating the highest detection accuracy (mAP50: 0.918, F1 score: 0.851). For semantic segmentation models, SegFormer showed the strongest performance (IoU of 0.662 for red-brown and 0.542 for yellow) but at the cost of longest training time, while EVitNet offered the most cost-effective solution achieving comparable accuracy to SegFormer but with a superior training efficiency with its lighter architecture. The accurate identification and symptom classification of crown damage symptoms support the calibration and validation of satellite-based monitoring systems and assist in the prioritisation of ground-based diagnosis or management interventions. Full article

The EN-HERITAGE project aims to define and prototype an integrated digital platform for the management of virtual models of buildings belonging to the historic built heritage, with a particular focus on slate roofing systems. The platform integrates IoT technologies for environmental monitoring, architectural surveys carried out using laser scanning and photogrammetry, HBIM models, and artificial intelligence algorithms for the analysis of degradation phenomena. The pilot application was conducted on the Albergo dei Poveri complex in Genoa, providing a replicable methodology for the planned conservation of the historic built environment. Preliminary results highlight the effectiveness of the platform in integrating heterogeneous data, providing stakeholders involved in the management of extensive architectural heritage with concrete support for decision-making processes and greater efficiency in planning maintenance and restoration interventions on historic buildings. Full article

Modern Earth observation combines high spatial resolution, wide swath, and dense temporal sampling, producing image grids and sequences far beyond the regime of standard vision benchmarks. Convolutional networks remain strong baselines but struggle to aggregate kilometre-scale context and long temporal dependencies without heavy tiling and downsampling, while Transformers incur quadratic costs in token count and often rely on aggressive patching or windowing. Recently proposed visual state-space models, typified by Mamba, offer linear-time sequence processing with selective recurrence and have therefore attracted rapid interest in remote sensing. This survey analyses how far that promise is realised in practice. We first review the theoretical substrates of state-space models and the role of scanning and serialization when mapping two- and three-dimensional EO data onto one-dimensional sequences. A taxonomy of scan paths and architectural hybrids is then developed, covering centre-focused and geometry-aware trajectories, CNN– and Transformer–Mamba backbones, and multimodal designs for hyperspectral, multisource fusion, segmentation, detection, restoration, and domain-specific scientific applications. Building on this evidence, we delineate the task regimes in which Mamba is empirically warranted—very long sequences, large tiles, or complex degradations—and those in which simpler operators or conventional attention remain competitive. Finally, we discuss green computing, numerical stability, and reproducibility, and outline directions for physics-informed state-space models and remote-sensing-specific foundation architectures. Overall, the survey argues that Mamba should be used as a targeted, scan-aware component in EO pipelines rather than a drop-in replacement for existing backbones, and aims to provide concrete design principles for future remote sensing research and operational practice. Full article

With the development of 6G technology, conventional wireless communication systems are increasingly unable to meet stringent performance requirements in complex and dynamic environments. Therefore, integrated sensing and communication (ISAC), which enables efficient spectrum sharing, has attracted growing attention as a promising solution. This paper provides a comprehensive survey of reconfigurable intelligent surface (RIS)-unmanned aerial vehicle (UAV)-assisted ISAC systems. It first introduces a four-dimensional quantitative evaluation framework grounded in information theory. Then, we provide a structured overview of coordination mechanisms between different types of RIS and UAV platforms within ISAC architectures. Furthermore, we analyze the application characteristics of various multiple access schemes in these systems. Finally, the main technical challenges and potential future research directions are discussed and analyzed. Full article

Background: Dietary habits established during young adulthood have long-term implications for health, and food choices among university students are strongly shaped by contextual factors. Institutional eating environments represent a relevant setting for promoting healthier dietary behaviors, yet limited evidence integrates students’ engagement with these settings, their food consumption patterns across contexts, and the individual decision-making processes underlying food choice. Methods: This cross-sectional study analyzed survey data from 1519 students enrolled at a large Italian university. Measures included sociodemographic characteristics, self-identified dietary style, engagement with the university canteen, consumption frequency of selected food categories across institutional and non-institutional contexts, and category-specific food-choice motivations. Data were analyzed using descriptive analyses, Borda count rankings, paired comparisons, and multiple linear regression models. Results: Clear contextual differences in food consumption emerged across all food categories, with consistently lower consumption frequencies within the university canteen compared to outside settings (all p < 0.001). The largest contextual gap was observed for fruit consumption (d = 0.94), with similarly pronounced differences for plant-based foods. Taste was the most salient decision-making factor across food categories (overall M ≈ 4.4), while health-related motives were more prominent for healthier foods and gratification for desserts. Across contexts, self-identified dietary style was the most consistent predictor of food consumption, explaining substantial variance for animal-based protein consumption (R² = 0.293 in the canteen; R² = 0.353 outside), whereas age and gender showed smaller, food-specific associations. Conclusions: The findings highlight institutional eating settings as distinct food environments in which individual dietary preferences are only partially expressed. Effective strategies to promote healthier eating among university students should move beyond generic approaches and integrate interventions targeting service-related engagement, category-specific choice architecture, and students’ dietary identities. Full article

Bio-electric fields—manifested as Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG), and Electrooculogram (EOG)—are fundamental to modern medical diagnostics but often suffer from severe data imbalance, scarcity, and environmental noise. Generative Adversarial Networks (GANs) offer a powerful, nonlinear solution to these modeling hurdles. This review presents a comprehensive survey of GAN methodologies specifically tailored for bio-electric signal processing. We first establish a theoretical foundation by detailing GAN principles, training mechanisms, and critical structural variants, including advancements in loss functions and conditional architectures. Subsequently, the paper extensively analyzes applications ranging from high-fidelity signal synthesis and noise reduction to multi-class classification. Special attention is given to clinical anomaly detection, specifically covering epilepsy, arrhythmia, depression, and sleep apnea. Furthermore, we explore emerging applications such as modal transformation, Brain–Computer Interfaces (BCI), de-identification for privacy, and signal reconstruction. Finally, we critically evaluate the computational trade-offs and stability issues inherent in current models. The study concludes by delineating prospective research avenues, emphasizing the necessity of interdisciplinary synergy to advance personalized medicine and intelligent diagnostic systems. Full article

The transition toward Industry 4.0 requires advanced sensing platforms capable of delivering real-time, high-fidelity data under extreme industrial conditions. Thin-film sensors, leveraging both photonic and functional approaches, are emerging as key enablers of this transformation. By exploiting optical phenomena such as Fabry–Pérot interference, guided-mode resonance, plasmonics, and photonic crystal effects, thin-film photonic devices provide highly sensitive, electromagnetic interference-immune, and remotely interrogated solutions for monitoring temperature, strain, and chemical environments. Complementarily, functional thin films including oxide-based chemiresistors, nanoparticle coatings, and flexible electronic skins extend sensing capabilities to diverse industrial contexts, from hazardous gas detection to structural health monitoring. This review surveys the fundamental optical principles, material platforms, and deposition strategies that underpin thin-film sensors, emphasizing advances in nanostructured oxides, 2D materials, hybrid perovskites, and additive manufacturing methods. Application-focused sections highlight their deployment in temperature and stress monitoring, chemical leakage detection, and industrial safety. Integration into Internet of Things (IoT) networks, cyber-physical systems, and photonic integrated circuits is examined, alongside challenges related to durability, reproducibility, and packaging. Future directions point to AI-driven signal processing, flexible and printable architectures, and autonomous self-calibration. Together, these developments position thin-film sensors as foundational technologies for intelligent, resilient, and adaptive manufacturing in Industry 4.0. Full article

Wireless Power Transfer (WPT) has become a pivotal technology, enabling the battery-free operation of Internet of Things (IoT) and biomedical devices while supporting environmental sustainability. This review provides a comprehensive analysis of microstrip patch antennas (MPAs) operating at the 5.8 GHz Industrial, Scientific, and Medical (ISM) band, emphasizing their advantages over the more commonly used 2.4 GHz band. A detailed and systematic classification framework for MPA architectures is introduced, covering single-element, multi-band, ultra-wideband, array, MIMO, wearable, and rectenna systems. The review examines advanced optimization methodologies, including Defected Ground Structures (DGS), Electromagnetic Bandgap (EBG) structures, Metamaterials (MTM), Machine Learning (ML), and nanomaterials, each contributing to improvements in gain, bandwidth, efficiency, and device miniaturization. Unlike previous surveys, this work offers a performance-benchmarked classification specifically for 5.8 GHz MPAs and provides a quantitative assessment of key trade-offs, such as efficiency versus substrate cost. The review also advocates for a shift toward Power Conversion Efficiency (PCE)-centric co-design strategies. The analysis identifies critical research gaps, particularly the ongoing disparity between simulated and experimental performance. The review concludes by recommending multi-objective optimization, integrated antenna-rectifier co-design to maximize PCE, and the use of advanced materials and computational intelligence to advance next-generation, high-efficiency 5.8 GHz WPT systems. Full article

Pollen identification is critical for melissopalynology (honey authentication), ecological monitoring, and allergen tracking, yet manual microscopic analysis remains labor-intensive, subjective, and error-prone when multiple grains overlap in realistic samples. Existing automated approaches often fail to address multi-grain scenarios or lack systematic comparison across classical and deep learning paradigms, limiting their practical deployment. This study proposes a subinstance-based classification framework combining YOLOv12n object detection for grain isolation, independent classification via classical machine learning (ML), convolutional neural networks (CNNs), or Vision Transformers (ViTs), and majority voting aggregation. Five classical classifiers with systematic feature selection, three CNN architectures (ResNet50, EfficientNet-B0, ConvNeXt-Tiny), and three ViT variants (ViT-B/16, ViT-B/32, ViT-L/16) are evaluated on four datasets (full images vs. isolated grains; raw vs. CLAHE-preprocessed) for four berry pollen species (Ribes nigrum, Ribes uva-crispa, Lonicera caerulea, and Amelanchier alnifolia). Stratified image-level splits ensure no data leakage, and explainable AI techniques (SHAP, Grad-CAM++, and gradient saliency) validate biological interpretability across all paradigms. Results demonstrate that grain isolation substantially improves classical ML performance (F1 from 0.83 to 0.91 on full images to 0.96–0.99 on isolated grains, +8–13 percentage points), while deep learning excels on both levels (CNNs: F1 = 1.000 on full images with CLAHE; ViTs: F1 = 0.99). At the instance level, all paradigms converge to near-perfect discrimination (F1 ≥ 0.96), indicating sufficient capture of morphological information. Majority voting aggregation provides +3–5% gains for classical methods but only +0.3–4.8% for deep models already near saturation. Explainable AI analysis confirms that models rely on biologically meaningful cues: blue channel moments and texture features for classical ML (SHAP), grain boundaries and exine ornamentation for CNNs (Grad-CAM++), and distributed attention across grain structures for ViTs (gradient saliency). Qualitative validation on 211 mixed-pollen images confirms robust generalization to realistic multi-species samples. The proposed framework (YOLOv12n + SVC/ResNet50 + majority voting) is practical for deployment in honey authentication, ecological surveys, and fine-grained biological image analysis. Full article

The reliable and continuous acquisition of seismic data from multiple open sources is essential for real-time monitoring, hazard assessment, and early-warning systems. However, the heterogeneity among existing data providers such as the United States Geological Survey, the European-Mediterranean Seismological Centre, and the Spanish National Geographic Institute creates significant challenges due to differences in formats, update frequencies, and access methods. To overcome these limitations, this paper presents a modular and automated framework for the scheduled near-real-time ingestion of global seismic data using open APIs and semi-structured web data. The system, implemented using a Docker-based architecture, automatically retrieves, harmonizes, and stores seismic information from heterogeneous sources at regular intervals using a cron-based scheduler. Data are standardized into a unified schema, validated to remove duplicates, and persisted in a relational database for downstream analytics and visualization. The proposed framework adheres to the FAIR data principles by ensuring that all seismic events are uniquely identifiable, source-traceable, and stored in interoperable formats. Its lightweight and containerized design enables deployment as a microservice within emerging data spaces and open environmental data infrastructures. Experimental validation was conducted using a two-phase evaluation. This evaluation consisted of a high-frequency 24 h stress test and a subsequent seven-day continuous deployment under steady-state conditions. The system maintained stable operation with 100% availability across all sources, successfully integrating 4533 newly published seismic events during the seven-day period and identifying 595 duplicated detections across providers. These results demonstrate that the framework provides a robust foundation for the automated integration of multi-source seismic catalogs. This integration supports the construction of more comprehensive and globally accessible earthquake datasets for research and near-real-time applications. By enabling automated and interoperable integration of seismic information from diverse providers, this approach supports the construction of more comprehensive and globally accessible earthquake catalogs, strengthening data-driven research and situational awareness across regions and institutions worldwide. Full article