Search Results (4,923)

Accurate and timely identification of slippery road surfaces is essential for ensuring driving safety and operational efficiency on highways. However, blurred vehicle-background boundaries, uneven illumination, and water splashing caused by passing vehicles make existing image-based recognition methods prone to low accuracy. To address these challenges, this paper proposes an improved SegFormer-based framework with two task-specific innovations: (1) a novel Guided Multi-scale Fusion (GMF) module for task-guided multi-scale feature integration, designed to incorporate auxiliary information such as vehicle type, relative speed, and splash regions, enabling the network to focus on slipperiness-relevant road areas while suppressing background interference; and (2) an improved Boundary Attention Awareness (BAA) module with directional Sobel-based boundary initialization, which provides explicit geometric priors to preserve fine boundary details and reduce ambiguity in slippery regions with irregular or weak edges. A multi-scale input and enhancement strategy is further adopted, along with a weighted combination of cross-entropy loss and Dice loss to mitigate class imbalance. Experimental results on our self-constructed Guangzhou Beierhuan Expressway dataset achieve an mIoU of 95.80%, accuracy of 97.84%, and F1-score of 97.86%. To verify cross-domain generalization, we further evaluate the model on two additional benchmarks: it achieves an mIoU of 93.51% on the synthetic SYN-UDTIRI dataset, and attains an mIoU of 95.80% with an AmIoU of 76.20% on the public Cityscapes dataset, achieving competitive performance against several mainstream architectures. The proposed method offers considerable application potential for highway safety warning systems. Full article

Real-time video anomaly detection (VAD) under realistic edge constraints—sub-second latency, ≤25 W power, no cloud dependency, and human-interpretable output—remains an open problem. Existing lightweight video convolutional neural networks (X3D, MoViNets) are bound to closed-set training distributions, while recent vision–language-model-based VAD methods (LAVAD, VERA, Holmes-VAD) achieve 80–89% area under the curve (AUC) but rely on datacenter-grade GPUs and Chain-of-Thought (CoT) reasoning that pushes per-segment latency well above one second. This paper reframes the design target from peak accuracy to practical edge deployability and contributes two tightly coupled designs: (i) an edge-optimized inference stack that compresses Qwen3-VL-2B with 4-bit Activation-aware Weight Quantization (INT4 AWQ) and serves it through a TensorRT-LLM C++ runtime on NVIDIA Jetson Orin NX (16 GB, 25 W); and (ii) a fully automatic, CoT-free verbalized prompt optimization in which an 8B optimizer iteratively refines a natural-language definition block $D_t$ using class-balanced (stratified) development batches on a disjoint development subset, with no human editing and no runtime cost on the edge device. Three findings support this framing: (a) the inference stack reduces per-segment latency to 0.25 s, a 7.4× speed-up and 55% memory reduction over a Python/PyTorch baseline; (b) verbalized prompt optimization improves zero-shot AUC from 71.82% (manual prompt) to 76.39%, outperforming GPT-4- and Gemini-Pro-generated prompts (74.12% and 74.35%) under the same edge backbone; and (c) single-frame input attains the highest mean AUC among one-, five-, and eight-frame windows—statistically comparable to the five-frame setting—while offering the lowest latency, making it the preferred operating point under the edge budget. While the absolute AUC (76.39%) is below recent server-side methods (CLIP-TSA 87.58%, VadCLIP 88.02%, Holmes-VAD 89.51%), our framework is the only one in this comparison that operates entirely on a ≤25 W edge device, providing a deployment-oriented operating point on the accuracy–feasibility frontier of VLM-based VAD. Full article

Urban grey infrastructure, including transport corridors, flood-control systems, drainage facilities, railway margins, and industrial brownfields, is often treated in conventional planning as functionally vacant or ecologically marginal space. In this study, grey infrastructure is defined as engineered or infrastructure-dominated urban spaces and their residual or edge landscapes, rather than as green infrastructure. Yet such spaces may still contain spontaneous vegetation, water edges, soil patches, and sheltered microhabitats that can support birds, insects, small mammals, and other urban species in highly urbanized environments. From a multispecies justice perspective, this study examines spatial mismatches between PPGIS-derived public recognition and habitat quality modeled using the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model in urban grey infrastructure. Using Zhanggong District, Ganzhou, China, as a case study, we developed an integrated framework that combines public participation geographic information systems (PPGIS), grid-based public recognition mapping, and the InVEST Habitat Quality model. Based on 260 valid questionnaires, participants identified locations associated with aesthetic preference and perceived habitat potential. These mapped perception points were converted into a grid-based public recognition indicator and compared with modeled habitat quality. The results indicate a marked perception-habitat mismatch. High public recognition was frequently associated with visually ordered, accessible, and highly managed urban spaces, whereas high modeled habitat quality was more strongly related to vegetated, riverine, peripheral, and semi-natural spaces. The spatial classification shows that aesthetic mismatch areas, where public recognition was high but modeled habitat quality was low, covered 43.20 km², accounting for 7.16% of the study area, whereas hidden ecological value areas, where modeled habitat quality was high but public recognition was low, covered 118.21 km², accounting for 19.59%. This study provides a replicable framework for identifying perception–habitat mismatches and supports more ecologically informed and inclusive strategies for regenerating urban grey infrastructure. More specifically, the study extends an established perception-ecology debate to grey infrastructure and residual urban spaces and demonstrates how PPGIS-derived public recognition and InVEST-modeled habitat quality can be compared within the same grid framework to locate spatially explicit alignment and mismatch. Full article

Timely forecasting of maize productivity is essential to support precision agriculture and optimize management practices. In this study, we analyzed the potential of integrating ground-based measurements and UAV-derived spectral data for predicting maize grain yield (GY) under different fertilization conditions. Field data were collected at two key phenological stages: early vegetative stage (V7) and pre-harvest (R4). Ground-based measurements included SPAD, above-ground biomass (AGB), and leaf area index (LAI), while multispectral and hyperspectral imagery was acquired by drone. A series of Ordinary Least Squares (OLS) models was developed to evaluate the predictive performance of individual variables and their combinations. Model robustness was assessed using two validation strategies: Leave-One-Treatment-Out (LOTO) to assess model performance across the treatments included in the experimental design and random sampling to assess performance within the dataset. The results showed that yield prediction was less accurate during the early growth stages, where data fusion significantly improved the model’s accuracy (${\mathrm{R}}^2$ = 0.82; MAE = 6.36 q ${\mathrm{ha}}^{-1}$; $MAPE\approx$7 %). The predictive performance of VIs alone increased substantially in the pre-harvest stage, with the combination of red-edge indices and LAI proving to be the best model for late yield prediction (${\mathrm{R}}^2$ = 0.86; MAE = 6.56 q ${\mathrm{ha}}^{-1}$; $MAPE\approx$7%). Comparison of multispectral and hyperspectral data revealed comparable predictive performance, suggesting that multispectral sensors may already capture the key spectral information needed for yield forecasting. Furthermore, random validation consistently produced more optimistic results than the LOTO method, highlighting the importance of using validation strategies that explicitly account for the experimental design when evaluating model performance across the treatments included in the study. Overall, the present study demonstrates that yield prediction is highly dependent on the phenological stage and validation approach, and that integrating complementary data sources can improve model performance, particularly during the early growth stages. These findings should be interpreted as a proof-of-concept based on a single-site, single-season experiment with a limited sample size (n = 12), and therefore require further validation across multiple environments and growing seasons. Full article

Real-time vehicle warning systems are critical for collision prevention, yet they face stringent sub-10 ms latency requirements under severe energy and computational constraints. This review systematically surveys low-latency edge computing architectures for such systems, explicitly comparing CPU-based, GPU-accelerated, FPGA-based, ASIC/NPU-embedded, fog, and cloud-only processing paradigms. We examine edge intelligence frameworks for intelligent transportation systems, the computational demands of collision avoidance algorithms, V2X communication protocols, and hardware accelerators. A key contribution is a comparative analysis of latency, power consumption, and area trade-offs, revealing that FPGA accelerators achieve deterministic sub-millisecond processing at 5–15 W, while emerging NPUs offer 1–5 W alternatives for fixed-function inference. A critical synthesis of the literature identifies major gaps: the absence of standardized benchmarks, insufficient field-testing of FPGA prototypes, and underutilized potential of approximate computing in safety loops. Furthermore, we introduce fog computing as a vital intermediary layer to bridge edge-cloud gaps. This review consolidates over 58 core studies and offers practical, actionable insights for designing next-generation vehicular safety systems. Full article

To address ecological security challenges in karst urban agglomerations, this study proposes a green infrastructure network (GIN) optimization framework that integrates geological characteristics with complex network theory. A fracture-modified minimum cumulative resistance (FM-MCR) model, coupled with an ant colony algorithm, was developed to enhance GIN extraction accuracy. Using a cascade failure model, the structural response of the GIN under simulated attack scenarios was systematically evaluated across four edge enhancement strategies, enabling spatial layout optimization. Network clustering and node centrality assessments were further applied to identify critical corridors and strategic nodes. Results identified 108 ecological sources, with Qiannan contributing the largest area (3141.97 km², 23.96% of the total), and 162 corridors in the original GIN. Among the four edge enhancement strategies, the low-degree-first (LDF) strategy significantly improved network robustness, decreasing percolation thresholds by 20.9%, 22.98%, and 16.99% under random, degree, and betweenness attacks, respectively. Network cluster analysis further delineated 45 GIN clusters, 19 critical corridors, and 11 strategic nodes, revealing cross-scale ecological hubs linking the Wumengshan–Daloushan corridor with Guiyang’s urban green wedge. This framework, characterized by geological correction, dynamic edge enhancement, and cluster-based management, offers a scientific basis for improving GIN resilience in karst regions. Full article

Clothing pressure is a critical design parameter in compression garments, yet its estimation in three-dimensional (3D) digital environments remains challenging because it depends on fabric mechanics, garment deformation, body geometry, and garment–body contact. This study developed a calibrated 3D vector-projection method for estimating clothing pressure from digital garment-mesh deformation. The method was based on the mechanical premise that garment extension generates in-plane tensile forces, whereas interface pressure is associated with the component of those forces acting normal to the body surface. Accordingly, corresponding flat and deformed garment meshes from CLO 3D were used to calculate edge-length strain and internal force; resultant forces were projected onto local avatar-normal directions and normalized by vertex-associated surface area to obtain uncalibrated pressure-related values. Five tricot fabrics and two pattern-reduction levels were used to produce ten compression tops, and pressure measured at five body locations was used for modulus-group-specific linear calibration to account for stiffness-dependent differences in deformation-to-pressure conversion. Under leave-one-garment-out cross-validation, the final linear model achieved an overall R² of 0.563, an RMSE of 0.587 kPa, and an MAE of 0.427 kPa. A 1–15 mm contact-distance analysis identified 10 mm as a conservative numerical stabilization point, with normalized means remaining within ±2% of the 15 mm reference and adjacent-threshold changes below 0.1 from 10 to 15 mm. The proposed method provides a transparent, mechanics-informed mesh-level procedure that converts digital garment deformation into calibrated body-normal pressure estimates and 3D spatial maps without treating commercial virtual-fitting pressure maps as direct physical predictions. Full article

In this paper, we evaluated image quality, algorithms and a workflow associated with matching horizons derived from 3D terrain data to ground imagery as a way to visually estimate a geographic position. The evaluation represented a passive, vision-based navigation technique using long-standoff (>1 KM) terrain features and a horizon detection algorithm hosted within an Android-based geospatial application. Our method involved quantitatively grading edge image feature quality based on pixel data between sky and terrain from high to poor. We tested the algorithm and processing to derive a position using images of varied quality, representing fine and gross, and near and far geographic terrain structures. The site chosen for our tests was located near the Organ Mountains in New Mexico to take advantage of largely unobstructed, long-distance features that challenged both image quality and horizon detection. Testing used the Samsung S23 Ultra (S23U) phone’s primary internal camera to acquire the necessary ground images and native compute power. Our evaluation workflow featured both pre-processing and near-real-time processing elements for position estimations. Pre-processing involved building a Geopackage containing geolocated, synthetic horizons extracted from available 3D terrain data of the test area and camera/sensor configuration data. These data were pre-loaded onto the phone to accomplish the live, near-real-time positional determinations matched to the extracted horizons generated from images acquired by the S23U camera. Our results showed that single-image processing, where only one high-quality ground photo was acquired, 75% of solutions were within 100 m of the actual camera position (compared with the internal sensor-based, Exchangeable Image File Format (EXIF) metadata). Single images of fair quality resulted in positional accuracies where only 38% of the solutions were within 100 m of the EXIF. Improvement was realized when four or more images from varying directions collected from a single location resulted in over 90% of positions falling within 100 m of the EXIF. Full article

Mainland Portugal has a lichenological tradition spanning more than three centuries, yet it still lacks a modern, critically revised checklist of its lichen biota. This gap is particularly relevant given the increasing role of species checklists as authoritative references for DNA metabarcoding applications and the expansion of large biodiversity datasets, including those derived from citizen science initiatives. Here we present the first comprehensive checklist of lichens and lichenicolous fungi recorded from Mainland Portugal, based on the critical assessment of 43,718 bibliographic records drawn from 381 publications issued between 1788 and 2021. A total of 2012 species and infraspecific taxa are documented from a universe of 4440 taxonomic names, including synonyms. Nomenclature was updated in accordance with the ITALIC 8.0 database. Species distributions are provided by province, following the scheme adopted in Portuguese botanical floras, with additional information on substrates and altitudinal ranges where available. With over 2000 taxa recorded, Mainland Portugal can be considered a lichen-rich territory relative to its land area, comparing favourably with other European checklists. We discuss the factors underlying this diversity, with particular emphasis on the biogeographical position of Portugal at the south-western edge of the European continent. Full article

Building materials and their spatial distribution play a significant role in determining the outcomes of human-caused and natural disasters in urban and peri-urban areas. However, building-level data on building and roofing materials are scarce. Here, we explore the feasibility and performance of a Convolutional Neural Network (CNN) model using spectrally transformed high-resolution multispectral imagery to map roofprints (i.e., classifying and delineating roofing materials at the building footprint-level) in Washington, District of Columbia (D.C.) and Denver, CO, United States. To generate consistent training data, we integrate geospatial vector data of individual building footprints with real estate industry-derived building-level roofing material data to create labeled image data from Planet SuperDove imagery. We compare the CNN classifier to a pixel-based machine learning (ML) model to demonstrate the capability of our roofprints mapping approach. With F1-scores ranging from 0.56 to 0.95 for the most common roof material classes, the CNN model outperformed the pixel-based ML classifier by 15% and 17% in Washington, D.C., and Denver, respectively. Results demonstrate within-domain robustness for the studied metropolitan areas, which are characterized by differing building densities, roof morphologies, and material patterns. While cross-region transferability was not evaluated, our findings provide a controlled comparison of pixel-based and context-aware approaches for rooftop material mapping and highlight the importance of hierarchical representations that integrate spectral information with roof texture, edge characteristics, spatial arrangement, and neighborhood context for improving classification performance. Accurately mapping building materials has the potential to advance urban planning and environmental policies, including assessments of heat exposure, energy demand, as well as hazard risk and community resilience. Full article

During the service of solid rocket motors, propellant grains need to bear various loads such as curing cooling, gravity, and combustion internal pressure. The internal pore defects will seriously affect the structural integrity. In this paper, a three-dimensional finite element model of propellant–insulation layer–mold is established to study the structural responses of pore defects with different sizes (30–100 mm) under three typical working conditions: curing cooling, curing cooling coupled with gravity, and internal pressure loading. It is found that under the curing cooling condition, compared with the non-porous propellant grain structure, the structure with pores will raise the overall mechanical response of the propellant grain, and the maximum stress and strain are mainly concentrated in the front end of the core hole and the wing groove area. The pore size has a limited impact on the overall stress distribution, but will change the local stress concentration degree. Among them, the 80 mm pore reduces the stress in the wing groove area through stress field interference. Moreover, large-size pores will significantly weaken the structural bearing capacity and increase the contact pressure between the propellant and the core mold. Under the condition of curing cooling coupled with gravity, the stress and strain are mainly distributed at the edge of the pores, and the values increase with the increase of pore size. Under the action of internal pressure load, the stress and strain in the middle section of the propellant grain have no obvious change, but stress concentration occurs in the transition area between the core hole and the wing groove and at the end of the wing groove. The results of this study provide a reference for the integrity evaluation and structural optimization of propellant grains with pore defects. Full article

Urban pluvial flood analysis requires spatial inundation information within operationally useful computation times. This study develops a graphics processing unit (GPU)-accelerated Node–Edge Urban Flood Model (GNE-UFM) that couples SWMM-style runoff generation, structured-grid surface flow, a typed sewer graph, and conservative surface–sewer exchange. The model was evaluated in the Sillim drainage district using the August 2022 observed event and three one-hour rainfall scenarios of 40, 90, and 150 mm, with inputs and output thresholds matched to InfoWorks ICM. For the synthetic scenarios, GNE-UFM achieved CSI values of 0.878–0.904 and wet-union RMSE values of 0.050–0.078 m, while full coupled GPU runs closed the effective runoff mass balance with absolute residuals no larger than 0.0284%. Runtime was comparable to ICM for the lowest-intensity case and became more efficient for the 90 and 150 mm cases, with GNE-UFM maintaining nearly constant RTR as rainfall intensity and inundated area increased. Full article

Real-time road accident prediction under dynamic climatic conditions remains a critical challenge for intelligent transportation systems, especially in peripheral and rural regions with limited communication infrastructure. This study proposes and evaluates a comprehensive five-layer cyber–physical architecture based on Federated Edge Intelligence to enable climate-aware spatiotemporal road accident prediction across the 24 provinces of Ecuador. The framework integrates low-power IoT sensing nodes equipped with TinyML capabilities (ESP32-S3), long-range LoRaWAN (Long-Range Wide-Area Network) communication networks, containerized edge–cloud orchestration via OpenNebula and K3s, a decentralized Federated Learning ecosystem using the FedAvg algorithm, and a geospatial decision intelligence backend. Leveraging a nationwide multi-source dataset spanning the 2014–2025 period with 27,620 processed records, the architecture successfully handles highly skewed historical accident profiles optimized through a Box–Cox transformation. Empirical results demonstrate that the centralized Stacking ensemble achieves the highest overall baseline performance ($R^2=0.2460,\mathrm{MAE}=0.4748$) in the Box–Cox transformed space. In the decentralized environment, the federated Gradient Boosting implementation establishes a resilient and viable accuracy trade-off ($14.51\%$ increase in MAE) while strictly maintaining localized data sovereignty and compliance with personal data protection legislation. Operationally, the edge nodes achieve a localized inference latency of only $78\mathrm{ms}$, well below the critical $200\mathrm{ms}$ safety threshold, while the global aggregation engine exhibits rapid convergence within just three communication rounds. This cyber–physical ecosystem proves that combining localized TinyML inference with federated aggregation provides a scalable, low-latency, and privacy-preserving foundation for next-generation climate-aware road safety infrastructures in connectivity-constrained environments. Full article

Testing advanced driver assistance systems (ADAS) under rare or safety-critical conditions is impractical at full scale: track campaigns are expensive, time-intensive, and cannot easily reproduce low-probability events. Scaled cyber–physical testbeds offer a more accessible path by coupling miniature vehicle platforms with virtual simulation environments, but the field has no unified review that covers co-simulation architectures, indoor localization, and ground truth strategies in a single treatment. This paper addresses that gap with a PRISMA 2020-compliant systematic review of 92 primary sources selected from 984 records identified across IEEE Xplore and Scopus. Three topic areas are examined: real-time co-simulation architectures built on AirSim, CARLA, Gazebo, and LGSVL, compared for ROS 2 integration, synchronisation model, and edge hardware suitability; three indoor localization methods, namely AprilTag fiducial tracking, Visual Simultaneous Localization and Mapping (VSLAM), and Ultra-Wideband (UWB) radio positioning, evaluated against shared accuracy, latency, infrastructure, and robustness criteria; and existing ground truth strategies for indoor localization benchmarking. A consistent finding across the corpus is that no controlled cross-method localization comparison exists for scaled testbeds. To address this, we introduce the Programmable Ground Truth Reference System (PGTRS), which renders spatial references on a programmable LED floor panel at a pixel pitch of approximately 3.9 mm, targeting sub-centimetre ground truth accuracy without dedicated motion-capture infrastructure. The concept is demonstrated within a 1:14 scale Digital Driving Twin (DDT) testbed built at the University of Applied Sciences Wildau at a hardware cost of approximately €6576. Design guidelines and open research challenges are discussed. Full article

Urban fringe forests deliver critical ecosystem services yet face irreversible loss and complex degradation under urbanization, while their fragmentation dynamics relative to core forests remain largely unquantified owing to the lack of a spatiotemporally consistent mapping approach. To address this gap, a multi-source remote sensing framework integrating land cover, population, nightlight, and land surface temperature data was developed to delineate urban fringe boundaries and quantify forest dynamics in Zhejiang Province via area-weighted centroids and landscape metrics, where high forest cover and a polycentric urban structure create a highly heterogeneous and dynamic fringe environment, enabling separate quantification of spatiotemporal fragmentation patterns for urban fringe and core forests from 2004 to 2024. The results are as follows: (1) The four dimensions (land, population, economy, and environment) produced spatiotemporally distinct boundaries, with multi-dimensional integration outperforming any single indicator and nighttime light being the best. Meanwhile, the total fringe area grew from 2989.04 to 3990.66 km² over two decades, with the fastest growth in 2004–2014 and the most rapid boundary shifts in 2014–2019. (2) The fringe forest proportion dropped from 24.72% to 18.37%, with the largest decline in southwestern high-forest cities. Meanwhile, Hangzhou and Ningbo fringe forests increasingly assumed the main ecological carrier role formerly held by core forests, with their centroids moved southwestward most markedly in 2009–2014 and displaying a more consistent direction than core forests. (3) Fragmentation metrics showed a higher patch density and splitting index but lower connectivity in the fringe than in the core, with intensification peaking in Zhoushan and coinciding with intensive edge expansion in 2009–2014, followed by later responses in the core. This study provides a transferable multi-dimensional remote sensing methodology for urban fringe mapping indicating that fringe forests may serve as early-warning signals of urbanization-induced forest degradation, enabling spatially targeted forest management across varied urban contexts. Full article