Search Results (8,617)

Tax evasion is a dynamic process reflecting continuous interaction between taxpayers and regulatory institutions rather than a static deviation from fiscal equilibrium. This study introduces a predator-prey model of tax evasion that translates the Lotka-Volterra framework from biology into budgetary dynamics. The model captures the feedback between the volume of tax evasion and the intensity of regulation, incorporating nonlinearity, implicit reactive lag, and adaptive response. Theoretical derivation and numerical simulation identify three dynamic regimes—stable equilibrium, limit-cycle oscillation, and instability—that arise through a Hopf bifurcation. Bifurcation maps in the (r, a), (r, b), and (r, c) parameter spaces reveal how control efficiency, institutional inertia, and behavioral feedback jointly determine fiscal stability. Results show that excessive enforcement may destabilize the system by inducing regulatory fatigue, while weak control enables exponential growth in evasion. The model provides a dynamic analytical tool for evaluating fiscal policy efficiency and identifying stability thresholds. Its findings suggest that adaptive, feedback-based regulation is essential for maintaining long-term tax discipline. The study contributes to closing the research gap by providing a unified dynamic framework linking micro-behavioral decision-making with macro-fiscal stability, offering a foundation for future empirical calibration and behavioral extensions of fiscal systems. Full article

Traditional methods for assessing sunflower yield across large agricultural fields are typically labor-intensive and time-consuming. This study explores the integration of unmanned aerial vehicle (UAV) imagery and the YOLOv11 deep learning model for automated sunflower head detection and yield estimation. Aerial imagery was collected from sunflower fields using UAVs, and a YOLOv11-based detection model was developed to identify sunflower heads efficiently. Model performance was optimized by tuning the Confidence Threshold and Intersection over Union (IoU) parameters. A total of 1290 image tiles derived from 215 UAV images were used for model training and evaluation. The dataset was divided into training and testing subsets with an 80:20 ratio. The optimal configuration, achieved at a Confidence Threshold of 0.50 and an IoU Threshold of 0.40, yielded balanced and accurate results, including a Precision of 0.84, Recall of 0.95, mAP@0.5 of 0.95, and an F1-score of 0.90. The findings demonstrate that parameter adjustment directly influences model detection accuracy and reliability. Overall, this study confirms that combining UAV remote sensing with YOLOv11 offers a robust and scalable approach for automated sunflower yield estimation, significantly reducing manual effort and processing time. Moreover, the proposed framework can be adapted for other high-value crops, contributing to the advancement of intelligent and data-driven agricultural management systems. Full article

Traditional plant scouting is often a costly and labor-intensive task that requires experienced specialists to diagnose and manage plant stresses. Artificial intelligence (AI), particularly deep learning and computer vision, offers the potential to transform scouting by enabling rapid, non-intrusive detection and classification of early stress symptoms from plant images. However, deep learning models are often opaque, relying on millions of parameters to extract complex nonlinear features that are not interpretable by growers. Recently, eXplainable AI (XAI) techniques have been used to identify key spatial regions that contribute to model predictions. This project explored the potential of convolutional neural networks (CNNs) for classifying the severity of chilli thrips damage in strawberry plants in Florida and employed XAI techniques to interpret model decisions and identify symptom-relevant canopy features. Four CNN architectures, YOLOv11, EfficientNetV2, Xception, and MobileNetV3, were trained and evaluated using 2353 square RGB canopy images of different sizes (256, 480, 640 and 1024 pixels) to classify symptoms as healthy, moderate, or severe. Trade-offs between image size, model parameter count, inference speed, and accuracy were examined in determining the best-performing model. The models achieved accuracies ranging from 77% to 85% with inference times of 5.7 to 262.3 ms, demonstrating strong potential for real-time pest severity estimation. Gradient-Weighted Class Activation Mapping (Grad-CAM) visualization revealed that model attention focused on biologically relevant regions such as fruits, stems, leaf edges, leaf surfaces, and dying leaves, areas commonly affected by chilli thrips. Subsequent analysis showed that model attention spread from localized regions in healthy plants to wide diffuse regions in severe plants. This alignment between model attention and expert scouting logic suggests that CNNs internalize symptom-specific visual cues and can reliably classify pest-induced plant stress. Full article

The current study investigates the influence of intersubject variability in ocular characteristics on the mapping of visual field (VF) sites to the pointwise directional angles in retinal nerve fiber layer (RNFL) bundle traces. In addition, the performance efficacy on the mapping of VF sites to the optic nerve head (ONH) was compared to ground truth baselines. Fundus photographs of 546 eyes of 546 healthy subjects (with no history of ocular disease or diabetic retinopathy) were enhanced digitally and RNFL bundle traces were segmented based on the Personalized Estimated Segmentation (PES) algorithm’s core technique. A 24-2 VF grid pattern was overlaid onto the photographs in order to relate VF test points to intersecting RNFL bundles. The PES algorithm effectively traced RNFL bundles in fundus images, achieving an average accuracy of 97.6% relative to the Jansonius map through the application of 10th-order Bezier curves. The PES algorithm assembled an average of 4726 RNFL bundles per fundus image based on 4975 sampling points, obtaining a total of 2,580,505 RNFL bundles based on 2,716,321 sampling points. The influence of ocular parameters could be evaluated for 34 out of 52 VF locations. The ONH-fovea angle and the ONH position in relation to the fovea were the most prominent predictors for variations in the mapping of retinal locations to the pointwise directional angle (p < 0.001). The variation explained by the model (R² value) ranges from 27.6% for visual field location 15 to 77.8% in location 22, with a mean of 56%. Significant individual variability was found in the mapping of VF sites to the ONH, with a mean standard deviation (95% limit) of 16.55° (median 17.68°) for 50 out of 52 VF locations, ranging from less than 1° to 44.05°. The mean entry angles differed from previous baselines by a range of less than 1° to 23.9° (average difference of 10.6° ± 5.53°), and RMSE of 11.94. Full article

Background/Objective: Bitewing radiographs are widely used for evaluating dental caries, restorations, and periodontal status due to their low radiation dose and high image quality. While artificial intelligence–based studies are common for other dental imaging modalities, multi-class diagnostic charting on bitewing radiographs remains limited. This study aimed to simultaneously detect eight periodontal and restorative parameters using a YOLOv8x-seg–based deep learning model and to assess its diagnostic performance. Materials and Methods: A total of 1197 digital bitewing radiographs were retrospectively analyzed and annotated by experts, resulting in 7860 labels across eight conditions. Periodontal conditions included alveolar bone loss, dental calculus, and furcation defects, while restorative and dental conditions comprised caries, cervical marginal gaps, open contacts, overhanging fillings, and secondary caries. The dataset was divided on a patient basis into training (80%), validation (10%), and test (10%) sets. The YOLOv8x-seg model was trained for 800 epochs with extensive data augmentation, and performance was evaluated using precision, recall, and F1-score, along with confusion matrices. Results: The model showed the highest accuracy in the alveolar bone loss class (precision: 0.84, recall: 0.93, F1: 0.88). While moderate performance was achieved for dental calculus (F1: 0.58) and caries (F1: 0.57) detection, lower scores were recorded in low-frequency classes such as cervical marginal gap (F1: 0.23), secondary caries (F1: 0.29), overhanging filling (F1: 0.35), furcation defect (F1: 0.40), and open contact (F1: 0.41). The overall segmentation performance achieved an mAP@0.5 of 0.30 and an mAP@0.5:0.95 of 0.10, indicating an acceptable performance level for segmentation-based multi-class models. Conclusions: The obtained findings demonstrate that the YOLOv8x-seg architecture can detect and segment periodontal conditions with high success and restorative parameters with moderate success in automation processes in bitewing radiographs. Accordingly, the model presents a methodologically feasible framework for the multiple and simultaneous radiographic evaluation of periodontal and restorative findings on bitewing radiographs, with performance varying across classes and lower sensitivity observed in low-frequency conditions. Full article

In this study, a hybrid sensor based on a defective square-truncated patch antenna (STPA) and a frequency-selective surface (FSS) was analyzed numerically and experimentally for different glucose–distilled water solutions. Here, an FSS was employed to enhance the sensitivity of the hybrid sensor. The sensing principle relies on monitoring variations in the loss tangent ($tan\delta$) and relative permittivity (${\epsilon }_r$) caused by different glucose concentrations applied to the sample under test (SUT). An open-ended coaxial probe was used to measure the complex permittivity of the solutions, which was then fitted to the Debye relaxation model. The simulated and experimental results of the novel sensor showed good agreement in a glucose concentration monitoring application. The sensor spanned the glucose range from 0 mg/dL to 5000 mg/dL, exhibiting a sensitivity of 55.44 kHz/mgdL⁻¹ and a figure of merit (FOM) of 6.23 × ${10}^{-4}$ (1/mgdL⁻¹) in the experiments and 53.60 kHz/mgdL⁻¹ and 1.71 × ${10}^{-4}$ (1/mgdL⁻¹) FOM in the simulations. When solutions with different concentrations were tested in the SUT, the resonance frequency of the antenna ($f_0$, in GHz) changed. To further characterize the sensor response, the relationship between the glucose concentration (C, in mg/dL) and $f_0$ was examined. A regression-based prediction model was constructed to map the measured scattering parameters to the glucose concentration, yielding a coefficient of determination ($R^2$) of 0.976. The high sensitivity, compact size, and compatibility with planar fabrication suggest that the proposed hybrid sensor has the potential to contribute to the development of non-invasive glucose-monitoring systems. Full article

To address the high-dilution issues arising from efficient mining methods such as medium-deep drilling for underground thin veins of precious metals, detecting raw rock fragments after blasting for subsequent sorting has become a cutting-edge research focus. With the continuous advancement of artificial intelligence, deep learning offers novel applications for rock detection. Accordingly, this study employs an improved lightweight YOLOv8n model to detect two typical thin-vein precious metal ores: gold ore and wolframite. In consideration of the computational resource constraints in underground environments, a triple optimization strategy is proposed. First, GhostConv and C2f-Ghost modules were introduced into the backbone network to reduce redundant computations while preserving feature representation capabilities. Second, the VoVGSCSP module was incorporated into the neck to further decrease model parameters and computational load. Finally, the ECA mechanism was embedded before the SPPF pooling layer to enhance feature extraction for ores and rocks, thereby improving detection accuracy. The results demonstrate that the GVE-YOLOv8 model contains only 2.28 million parameters—a 24.3% reduction compared to the original YOLOv8n. FLOPs decrease from 8.1 G to 5.6 G, and the model size reduces from 6.3 MB to 4.9 MB, while detection accuracy improves to 98.3% mAP50 and 95.3% mAP50-95. This enhanced model meets the performance requirements for accurately detecting raw ore and rock fragments after underground blasting, thereby providing a novel research method for thin-vein mining. Full article

Large language models (LLMs) demonstrate outstanding capability in understanding natural language and show great potential in open-domain travel planning. However, when confronted with multi-constraint itineraries, personalized recommendations, and scenarios requiring rigorous external information validation, pure LLM-based approaches lack rigorous planning ability and fine-grained personalization. To address these gaps, we propose the Symbolic LoRA Travel Planner (SLTP) framework—an agent architecture that combines a two-stage symbol-rule LoRA fine-tuning pipeline with a user multi-option heuristic tree search (MHTS) planner. SLTP decomposes the entire process of transforming natural language into executable code into two specialized, sequential LoRA experts: the first maps natural-language queries to symbolic constraints with high fidelity; the second compiles symbolic constraints into executable Python planning code. After reflective verification, the generated code serves as constraints and heuristic rules for an MHTS planner that preserves diversified top-K candidate itineraries and uses pruning plus heuristic strategies to maintain search-time performance. To overcome the scarcity of high-quality intermediate symbolic data, we adopt a teacher–student distillation approach: a strong teacher model generates high-fidelity symbolic constraints and executable code, which we use as hard targets to distill knowledge into an 8B-parameter Qwen3-8B student model via two-stage LoRA. On the ChinaTravel benchmark, SLTP using an 8B student achieves performance comparable to or surpassing that of other methods built on DeepSeek-V3 or GPT-4o as a backbone. Full article

The accurate retrieval of the raindrop size distribution (DSD) is a longstanding objective in meteorology because it underpins reliable quantitative precipitation estimation. Among remote sensors, weather radars are the primary tool for mapping DSD over wide areas, and phased-array systems in particular have demonstrated unique advantages owing to their high temporal and spatial resolution together with agile beam steering. Exploiting the underused high-resolution capability of an X-band phased-array radar, this study induced a Rainfall Regression Model (RRM). The RRM assumes a normalized gamma DSD model and retrieves its three parameters. It was then applied to a rain event influenced by the remnant circulation of Typhoon Haikui that affected Guangzhou on 8 September 2023. First, collocated disdrometer observations and T-matrix scattering simulations are used to build polynomial regressions between DSD parameters (D₀, N_w, μ) and the polarimetric variables. Validation against independent disdrometer samples yields Nash–Sutcliffe efficiencies of 0.93 for D₀ and 0.91 for log₁₀N_w. The RRM is then applied to the full volumetric radar data. Horizontal maps reveal that the surface elevation angle consistently exhibited the largest standard deviation for all three parameters. A vertical profile analysis shows that large-drop cores (D₀ > 2 mm) can reside above 2 km and that iso-value contours tilt rather than align vertically, implying an appreciable horizontal drift of raindrops within the complex remnant typhoon–monsoon wind field. By demonstrating the ability of X-band phased-array radar to resolve the three-dimensional microphysical structure of remnant typhoon precipitation, this study advances our understanding of the vertical characteristics of raindrops and provides high-resolution DSD information that can be directly ingested into severe weather monitoring and nowcasting systems. Full article

We present an adaptive governance engine for blockchain-enabled Vehicular Ad Hoc Networks (VANETs) that regulates the latency–energy–coherence trade-off under rapid topology changes. The core contribution is an Ideal Information Cycle (an operational abstraction of information injection/validation) and a modular VANET Engine implemented as a real-time control loop in NS-3.35. At runtime, the Engine monitors normalized Shannon entropies—informational entropy S over active transactions and spatial entropy $H_{\mathrm{spatial}}$ over occupancy bins (both on $[0,1]$)—and adapts the consensus mode (latency-feasible PoW versus signature/quorum-based modes such as PoS/FBA) together with rigor parameters via calibrated policy maps. Governance is formulated as a constrained operational objective that trades per-block resource expenditure (radio + cryptography) against a Quality-of-Information (QoI) proxy derived from delay/error tiers, while maintaining timeliness and ledger-coherence pressure. Cryptographic cost is traced through counted operations, $E_{\mathrm{crypto}}=e_h{n}_{\mathrm{hash}}+e_{\mathrm{sig}}{n}_{\mathrm{sig}}$, and coherence is tracked using the LCP-normalized definition $D_{\mathrm{ledger}}\left(t\right)$ computed from the longest common prefix (LCP) length across nodes. We evaluate the framework under urban/highway mobility, scheduled partitions, and bounded adversarial stressors (Sybil identities and Byzantine proposers), using 600 s runs with 30 matched random seeds per configuration and 95% bias-corrected and accelerated (BCa) bootstrap confidence intervals. In high-disorder regimes ($S\gtrsim 0.8$), the Engine reduces total per-block energy (radio + cryptography) by more than 90% relative to a fixed-parameter PoW baseline tuned to the same agreement latency target. A consensus-first triggering policy further lowers agreement latency and improves throughput compared with broadcast-first baselines. In the emphasized urban setting under high mobility ($v=30$ m/s), the Engine keeps agreement/commit latency in the sub-100 ms range while maintaining finality typically within sub-150 ms ranges, bounds orphaning (≤10%), and reduces average ledger divergence below $0.07$ at high spatial disorder. The main evaluation is limited to $N\le 100$ vehicles under full PHY/MAC fidelity. PoW targets are intentionally latency-feasible and are not intended to provide cryptocurrency-grade majority-hash security; operational security assumptions and mode transition safeguards are discussed in the manuscript. Full article

A key challenge in automated marigold harvesting lies in the accurate identification of picking points under complex environmental conditions, such as dense shading and intense illumination. To tackle this problem, this research proposes a lightweight instance segmentation model combined with a harvest position estimation method. Based on the YOLOv11n-seg segmentation framework, we develop a lightweight PDS-YOLO model through two key improvements: (1) structural pruning of the base model to reduce its parameter count, (2) incorporation of a Channel-wise Distillation (CWD)-based feature distillation method to compensate for the accuracy loss caused by pruning. The resulting lightweight segmentation model achieves a size of only 1.3 MB (22.8% of the base model) and a computational cost of 5 GFLOPs (49.02% of the base model). At the same time, it maintains high segmentation performance, with a precision of 93.6% and a mean average precision (mAP) of 96.7% for marigold segmentation. Furthermore, the proposed model demonstrates enhanced robustness under challenging scenarios including strong lighting, cloudy weather, and occlusion, improving the recall rate by 1.1% over the base model. Based on the segmentation results, a method for estimating marigold harvest positions using 3D point clouds is proposed. Fitting and deflection angle experiments confirm that the fitting errors are constrained within 3–12 mm, which lies within an acceptable range for automated harvesting. These results validate the capability of the proposed approach to accurately locate marigold harvest positions under top-down viewing conditions. The lightweight segmentation network and harvest position estimation method presented in this work offer effective technical support for selective harvesting of marigolds. Full article

To investigate the hot deformation characteristics of the Al–10Mg–3Zn alloy, a series of hot compression tests was carried out using a Gleeble-3500 simulator. The experimental matrix covered temperatures of 300–450 °C and strain rates from 0.001 to 10 s⁻¹. The true stress–strain curves were obtained and the hot processing map of the alloy was constructed based on the Dynamic Material Model principle. The multi-objective optimization of the extrusion process parameters was performed using the response surface method. The results showed that the flow stress of Al–10Mg–3Zn alloy increased with the increase in the strain rate and decreased with the increase in the deformation temperature, indicating that the alloy had a positive strain rate sensitivity. A strain-compensated Arrhenius constitutive model and a hot processing map of Al–10Mg–3Zn alloy were established based on the temperature-corrected data; here, the optimal temperature range and strain rate range for hot processing were specified. The optimal extrusion process parameters, determined by the response surface method, were as follows: billet temperature of 400 °C, extrusion speed of 0.20 mm/s, and ingot length of 350 mm. With this parameter combination, the simulation predicted an extrusion load of 73.29 MN, a velocity deviation of 24.96%, and a cross-sectional temperature difference of 9.48 °C for the profile. The predicted values from the response surface method were highly consistent with those from the finite element simulation. The optimized process parameters significantly reduced the extrusion load of the profile. Full article

Coastal lakes require monitoring approaches that capture spatial and temporal variability beyond the limits of conventional in situ measurements. In this study, a SIGMaL framework (Satellite–In situ–GIS-multicriteria decision analysis (MCDA)–Machine Learning (ML)) was developed, a unified methodology that integrates in situ monitoring, GIS MCDA-derived water quality index (WQI), satellite imagery, and ML models for comprehensive coastal lake water quality assessment. A WQI, derived from a 12-month series of in situ measurements and environmental parameters, was used alongside four physicochemical parameters measured by a multiparameter probe. First, satellite reflectance from each sensor was used to train a set of nine regression models for modelling electrical conductivity (EC), turbidity, water temperature (WT), and dissolved oxygen (DO). Second, convolutional neural networks (CNNs) with spectral and temporal inputs were trained to classify WQI classes, enabling a cross-sensor evaluation of their suitability for lake water quality monitoring. Third, the trained CNNs were applied to generate WQI maps for a subsequent 12-month period without in situ data. Across all analyses, WQI-based models provided more stable and accurate models than those trained on raw parameters. Sentinel-2 achieved the most consistent WQI performance (AUC ≈ 1.00, R² ≈ 0.84), PlanetScope captured fine-scale spatial detail (R² ≈ 0.77), while Landsat 8–9 was most effective for WT but less reliable for multi-class WQI discrimination. Sentinel-2 is recommended as the primary satellite sensor for WQI mapping within the SIGMaL framework. These findings demonstrate the advantages of WQI-based modelling and highlight the potential of ML–remote sensing integration to support coastal lake water quality monitoring. Full article

This study aims to investigate the tip resistance mechanism of open-ended steel pipe piles under partially plugged conditions by decomposing the load-sharing contribution of the ring zone and the internal soil core. A virtual static loading test was performed using the two-dimensional discrete element method (2D-DEM). Note that the findings of this study were obtained within the range of the 2D-DEM analysis conditions and do not intend to directly reproduce the three-dimensional arching mechanism or to establish equivalence between 2D and 3D responses. Quasi-static conditions were ensured by identifying loading parameters such that the energy residual remained ≤5% during driving, rest, and static loading phases, and the sensitivity criterion |Δq_b|/q_b ≤ 3% was satisfied when the loading rate was halved or doubled. The primary evaluation range of static loading was set to s/D = 0.1 (10% D), corresponding to the displacement criterion for confirming the tip resistance in the Japanese design specifications for highway bridges. For reference, the post-peak mechanism was additionally tracked up to s/D = 0.2 (20% D). Within a fixed evaluation window located immediately beneath the pile tip, high-contact-force (HCF) points were binarized using the threshold τ = μ + σ, and their occupancy ratio φ and normalized force intensity I* were calculated separately for the ring and core regions. A density-based contribution index (“K_{-density share}”) was defined by combining “strength × area” and normalizing by the geometric width. The results suggest that, for the sand conditions and particle-scale ratios examined (D/d_50 = 25–100), the ring zone tends to carry on the order of 85–90% of the tip resistance within the observed cases up to the ultimate state. Even at high plugging ratios (CRs), the internal soil core gradually increases its occupancy and intensity with settlement; however, high-contact-force struts beneath the ring remain active, and it is suggested that the ring-dominant load-transfer mechanism is generally preserved. In the post-peak plastic regime, the K_{-density share} remains around 60%, indicating that the internal core plays a secondary, confining role rather than becoming dominant. These findings suggest that the conventional plug/unplug classification based on PLR can be supplemented by a combined use of plugging ratio CR (a kinematic indicator) and the ring contribution index (K_{-density share}), potentially enabling a continuous interpretation of plugged and unplugged behaviors and contributing to the establishment of a design backbone for tip resistance evaluation. Calibration of design coefficients, scale regression, and mapping to practical indices such as N-values will be addressed in part II of this study. (Note: “Contribution” in this study refers to the HCF-based density contribution index K_{-density share}, not the reaction–force ratio.) Full article

Accurately predict the external corrosion rate is crucial for the integrity management and risk assessment of buried pipelines. However, existing prediction models often suffer from limitations such as low accuracy, instability, and overfitting. To address these challenges, this study proposes a novel hybrid model, FA-IDBO-KELM. Firstly, Factor Analysis (FA) was employed to reduce the dimensionality of ten original corrosion-influencing factors, extracting seven principal components to mitigate multicollinearity. Subsequently, the hyperparameters (penalty coefficient C and kernel parameter γ) of the Kernel Extreme Learning Machine (KELM) were optimized using an Improved Dung Beetle Optimizer (IDBO). The IDBO included four key enhancements compared to the standard DBO: spatial pyramid mapping (SPM) for population initialization, a spiral search strategy, Lévy flight, and an adaptive t-distribution mutation strategy to prevent premature convergence. The model was validated using a dataset from the West–East Gas Pipeline, with 90% of the data being used for training and 10% for testing. The results demonstrate the superior performance of FA-IDBO-KELM, which achieved a root mean square error (RMSE) of 0.0028, a mean absolute error (MAE) of 0.0021, and a coefficient of determination (R²) of 0.9954 on the test set. Compared to benchmark models (FA-KELM, FA-SSA-KELM, FA-DBO-KELM), the proposed model reduced the RMSE by 93.0%, 89.1%, and 85.3%, and improved the R² by 85.7%, 10.6%, and 7.4%, respectively. The FA-IDBO-KELM model provides a highly accurate and reliable tool for predicting the external corrosion rate, which can significantly support pipeline maintenance decision-making. Full article