Search Results (2,528)

Motion sickness is expected to become more prevalent in road transportation as passengers increasingly engage in non-driving-related activities under automated driving. This study quantified the relative contributions of exposure time (discrete trial index), physiological signals, and vibration-derived features to motion sickness prediction under controlled road vibration. Thirty-nine participants were exposed to four road profiles reproduced on a motion simulator while electrocardiography (ECG), galvanic skin response (GSR), triaxial head acceleration, and motion sickness ratings were recorded across ten 2 min trials per session during a reading task. Three dataset configurations were compared: exposure time only (type 1), physiological and vibration-derived features only (type 2), and the combined (type 3) dataset. Motion sickness prediction was evaluated using support vector classification, random forest, and XGBoost, and permutation importance was used to assess feature contribution. Across all classifiers, exposure time was ranked as the most important feature, followed by overall root-mean-square, while ECG and GSR contributed at lower levels. The combined dataset achieved the best performance, reaching 77% classification accuracy and R² values up to 0.49. Leave-one-subject-out (LOSO) cross-validation further confirmed the feature importance ranking under subject-level evaluation, while physiological features showed near-zero importance for unseen participants, suggesting their contributions are largely individual-specific. Full article

Commercial plots in hot–humid tropical island cities require effective pedestrian-level ventilation; however, the extent to which different layout forms enhance wind-driven cooling potential under fixed development intensity remains insufficiently quantified. Taking Haikou as a representative tropical island case, this study examines how building count and spatial enclosure form affect the pedestrian-level wind environment of a commercial plot. A total of 63 layouts with one, two, and three building units were constructed under identical development constraints. ANSYS Fluent 2023 R1 (ANSYS, Inc., Canonsburg, PA, USA). was used to simulate pedestrian-level wind fields under representative summer southerly and east–northeasterly (ENE) wind conditions. Six ventilation-related indicators, including area-weighted mean wind speed, maximum wind speed, and the non-low-wind-speed area ratio in both seasons, were integrated into a Wind-Driven Cooling Potential Index (WDCPI). The weighting scheme combined climate-informed seasonal weights with entropy-based objective indicator weights. The results show that summer ventilation is more sensitive to layout form than winter ventilation. Although the average WDCPI decreases as building subdivision increases, layout B2 shows the highest WDCPI among the tested scenarios because its open inter-building space forms a continuous ventilation path aligned with the prevailing summer wind. The sensitivity analysis supports the relative stability of the ranking results. These findings highlight airflow connectivity and windward openness as important layout-screening principles for cooling-oriented commercial plot design in tropical island cities. Full article

As an important infrastructure for the construction of a new power system, pumped-storage hydropower stations may exert certain impacts on the surrounding fractional vegetation cover (FVC) during their construction. Based on Landsat remote sensing imagery and China Land Cover Dataset (CLCD) land use data from 2013 to 2024, this study investigated the dynamic changes in FVC and the spatial extent of engineering disturbance associated with the Zhen’an Pumped-Storage Hydropower Station in Shaanxi Province, China. The analysis integrated the Pixel Dichotomy Model, Theil-Sen trend analysis, Mann–Kendall significance test, coefficient of variation, landscape pattern indices, and correlation analysis. The results showed that: (1) FVC in the study area exhibited distinct stage-dependent evolution characteristics that were highly consistent with the construction timeline of the project. (2) The spatial influence of engineering disturbance on FVC was mainly concentrated within 1250 m, with the 0–250 m zone identified as the core impact area. Landscape fragmentation in this zone was higher than in other distance ranges, and vegetation degradation gradually weakened with increasing distance from the project. (3) Land use change within the study area was primarily characterized by the conversion of forest to cropland and impervious surfaces, resulting in a reduction in high coverage vegetation areas. Landscape patterns exhibited pronounced buffer-gradient characteristics. Within 500 m, the Largest Patch Index (LPI) decreased while the Shannon Diversity Index (SHDI) increased, indicating weakened continuity of dominant landscape patches. Beyond 500 m, LPI generally increased and SHDI decreased, suggesting a trend toward a more stable landscape structure. (4) Both air temperature and precipitation exhibited interannual fluctuations, but neither showed a significant long-term trend. The correlations between climatic factors and FVC were relatively weak, and the multiple regression model demonstrated limited explanatory power for FVC variation. Full article

Digital transformation is changing agricultural information access, market connectivity, and factor allocation, but its implications for crop-planting structure remain underexamined. Using a prefecture-level panel of 291 Chinese cities from 2000 to 2023, this study investigates whether the digital economy is associated with relative non-grain cropping, measured as the share of non-grain crop sown area in total crop sown area. The results show that a one-standard-deviation increase in the digital economy index is positively associated with relative non-grain cropping. Area decomposition indicates that this relationship reflects selective contraction and factor reallocation rather than unrestricted absolute expansion of non-grain acreage. Moderation and heterogeneity analyses further show that platform-commerce and digital-payment environments strengthen the association, while major grain-producing area status and stronger agricultural resource endowments weaken it. Robustness and endogeneity-mitigation checks support the positive association but also indicate important identification boundaries. The findings contribute to debates on sustainable land-use transition by showing how digitalization can reshape crop allocation under market opportunities, factor constraints, and food-security institutions. Full article

Emulsified collectors (ECs) have attracted increasing attention in coal slime flotation because of their superior dispersion characteristics and collecting performance compared with conventional hydrocarbon collectors. However, the relationship between flotation performance and the effects of ECs on microorganisms in circulating water remains poorly understood. In this study, four ECs were prepared using cationic (dodecyltrimethylammonium bromide (DTAB)), anionic (sodium dodecyl sulfate (SDS)), nonionic (Tween-80), and solid particle-based (β-cyclodextrin (CD)) emulsifiers. The droplet size distribution, flotation performance, adsorption behavior, and microbial response were systematically evaluated. The results showed that ECs significantly improved coal slime flotation compared with conventional kerosene. Among the surfactant-based ECs, smaller oil droplet sizes resulted in higher clean coal recovery, demonstrating the critical role of collector dispersion in flotation performance. Fourier transform infrared spectroscopy provided qualitative evidence of stronger relative adsorption of ECs on coal surfaces than that of conventional kerosene. Yeast survival tests revealed that all ECs exhibited less adverse impact on microorganisms than kerosene. However, for surfactant-based ECs, microbial survival decreased as flotation performance increased. A preliminary negative correlation (R² = 0.9323, n = 4, where “n” denotes the number of collector formulations tested) was observed between the yeast survival rate and the number of oil droplets, suggesting that fine oil droplets remaining in the aqueous phase may have contributed to reduced microbial viability. In contrast, the β-CD-stabilized Pickering emulsion achieved both the highest flotation efficiency index (56.22) and the highest yeast survival rate (66.49%), outperforming Tween-80-EC (54.90 and 58.65%), SDS-EC (50.95 and 61.45%), DTAB-EC (50.08 and 63.31%), and conventional kerosene (40.32 and 53.35%). These findings demonstrate the importance of balancing flotation performance and environmental compatibility in the design of sustainable flotation collectors for coal processing. Full article

Background: First-trimester screening for preterm preeclampsia relies on validated competing-risks models integrating maternal characteristics with biochemical and biophysical markers to generate individualized risk estimates. Large language models (LLMs) are increasingly used for pregnancy-related health information, yet their alignment with established clinical risk models remains unclear. Objective: to perform an exploratory, local perturbation-based assessment of how LLM-generated numerical risk estimates reproduce the direction and relative magnitude of established preeclampsia predictor effects compared with the Fetal Medicine Foundation (FMF) competing-risks model. Methods: A total of 129 synthetic clinical scenarios were generated from a low-risk reference pregnancy using a one-factor-at-a-time perturbation approach across 16 risk factors represented as 22 predictors. Eight LLMs (Claude, GPT, DeepSeek, Gemini, Copilot, Meta, Mistral, and Grok) estimated the probability of preeclampsia requiring delivery before 37 weeks. Corresponding risks were calculated using the FMF model. Model behavior was analyzed in logit space using a local perturbation-based modeling approach inspired by Local Interpretable Model-Agnostic Explanations (LIME) to derive feature-effect coefficients. Agreement with the FMF model was assessed using correlation, directional concordance, cosine similarity, and normalized root mean squared error, summarized using an exploratory composite score. Results: The FMF model identified mean arterial pressure, placental growth factor, parity, uterine artery pulsatility index, and chronic hypertension as dominant predictors. Alignment between LLM outputs and the FMF model was heterogeneous, with composite scores ranging from 0.59 to 0.82. Models with higher descriptive scores preserved predictor directionality (up to 90.9%) and rank ordering, but agreement in magnitude and scaling was limited (R²: 0.44–0.59). Intermediate models showed preserved directionality with reduced magnitude agreement, while lower-scoring models demonstrated more frequent sign inconsistencies and minimal variance explained. Conclusions: LLMs demonstrated partial, prompt-specific alignment with the FMF model in this local perturbation analysis, particularly for predictor direction and relative importance, but did not consistently reproduce quantitative effect sizes. This approach was intended to characterize local model behavior around a predefined reference case rather than evaluate clinically realistic combinations of interacting risk factors or global clinical prediction performance. Given the evolving nature of LLMs, ongoing reassessment using standardized approaches is required. Full article

Reliable estimation of winter wheat yield is critical to food system stability and farmland management. Integrating multi-spectral remote sensing data with agronomic parameters represents a primary strategy for improving yield estimation accuracy. However, existing research often overlooks parameters reflecting crop population structure and fails to account for dynamic shifts in the contributions of multidimensional agronomic variables across growth stages, thereby limiting prediction accuracy and model stability. To address these limitations, a winter wheat yield estimation model was developed. This model integrates multi-source and multi-temporal data, incorporates stem tiller density, a key population structure parameter, and accounts for dynamic variation across growth stages. Unmanned aerial vehicle multi-spectral images were collected at four key growth stages: jointing (stem elongation with detectable nodes), booting (flag leaf sheath swelling preceding heading), heading (spike emergence) and filling (grain filling with dry matter accumulation). Three growth indicators, stem tiller density, leaf area index and above-ground biomass, were measured. Two comprehensive growth indicators were derived using the coefficient of variation and the CRITIC weighting methods (CGI_CR). Correlation and feature importance analyses were used to identify sensitive vegetation indices (VIs), which were subsequently integrated with the comprehensive growth indicators. Single-stage, multi-source feature fusion and multi-temporal yield estimation models were established using the Kernel Extreme Learning Machine and its optimised algorithm using the Crested Porcupine Optimizer. The results showed the following: (1) among the individual growth stages, features from the filling stage achieved the highest prediction accuracy; (2) the fusion of multi-source features (VIs + CGI_CR) enhanced the prediction accuracy of the model, achieving a validation set R² of 0.884 and a relative prediction deviation of 2.916 at the filling stage; and (3) the multi-temporal model further improved predictive performance, with the validation R² reaching 0.920, indicating that information from different growth stages contributed complementarily to yield prediction and improved overall model performance. By contrast, the model exhibited relatively weak predictive capability at the early growth stages and was better-suited to early risk identification. Meanwhile, its generalisation ability under cross-regional and inter-annual conditions still requires further validation. Overall, integrating multi-source and multi-temporal data can enhance the precision and stability of predicting winter wheat yield, thereby facilitating precision agriculture management. Full article

Water quality analysis plays an essential role in maintaining the health and sustainability of river ecosystems, especially in semi-arid regions like the Arroyo Colorado Watershed in South Texas. Since the river is a vital source of water supply for local communities, agriculture, and wildlife, it faces significant challenges and pollution from land use changes, climate variation, and agricultural runoff. Continuous monitoring and assessment of water quality parameters and their temporal variability are essential to ensure the drinking water supply and aquatic ecosystem health. However, comprehensive laboratory-based water quality investigations are often constrained by higher costs, logistical complexity, and limited manpower. As a result, monitoring datasets are often not available for all water quality parameters, or the datasets may be incomplete. To address such challenges, the objective of this study was to evaluate the potential of water quality index (WQI)-based assessment supported by machine learning algorithms as an alternative decision-support tool for water quality evaluation. The analysis compared four monitoring stations in the Austin and Arroyo Colorado Watersheds, with particular emphasis on one gauging station at Port Harlingen. Datasets were collected from the Texas Commission of Environmental Quality (TCEQ). A complete exploratory data analysis (EDA) was performed to understand the TCEQ water quality datasets containing sixteen parameters, and seven water quality parameters were selected based on multicollinearity checks. It was observed that seven independent water quality parameters (dissolved oxygen, ammonia, nitrate, phosphorus, temperature, fecal coliform, and residual non-filterable material concentrations) were identified as sufficient to define the WQI of the Austin monitoring stations. Moreover, U.S. Environmental Protection Agency (EPA)-based guidelines were utilized to scale individual parameters to a range of 0–100 to remove their magnitude and correlation-based bias. These parameters were further analyzed using machine learning techniques, i.e., principal component analysis, K-means, and one-class support vector machine, to compute the relative importance based on their fluctuation within the temporal dataset. Finally, the mean WQI model was developed for Port Harlingen and achieved a strong agreement with the National Sanitation Foundation (NSF) WQI (R² = 0.91). These findings demonstrate the applicability of the proposed data-driven WQI framework for regional water quality assessment and comparative analysis across watersheds. Full article

Background/Objectives: Perceived stress and sleep quality are increasingly recognized as important determinants of emotional eating (EE), particularly among university students who are frequently exposed to academic pressures, psychosocial challenges, and lifestyle disruptions. Both stress and sleep quality are closely interconnected, with elevated stress contributing to sleep disturbances and poor sleep further exacerbating psychological distress. The present study aims to simultaneously examine the independent contributions of perceived stress and sleep quality to EE while accounting for other relevant sociodemographic, lifestyle, and anthropometric factors. Methods: This cross-sectional study was conducted among 1297 university students recruited from 10 geographical regions of Greece. Data on sociodemographic characteristics, academic profile, lifestyle behaviors, and anthropometric parameters were collected using standardized and validated assessment instruments. Perceived stress was measured using the Perceived Stress Scale (PSS), while sleep quality was evaluated with the Pittsburgh Sleep Quality Index (PSQI). Emotional eating was assessed using the Emotional Eating subscale of the Three-Factor Eating Questionnaire–Revised 18 (TFEQ-R18), and participants were subsequently classified into tertiles corresponding to low, moderate, and high levels of emotional eating. Results: Perceived stress was independently associated with EE in the multivariable analysis. Compared with students reporting low levels of stress, those experiencing high perceived stress exhibited almost threefold greater odds of being classified in higher emotional eating categories (OR = 2.86; p = 0.0001). Likewise, participants with moderate perceived stress were approximately twice as likely to demonstrate elevated emotional eating levels relative to their low-stress counterparts (OR = 2.01; p = 0.0001). Sleep quality was also independently associated with EE after adjustment for potential confounding factors. Specifically, students characterized by inadequate sleep quality had more than three times the likelihood of belonging to higher emotional eating categories compared with those reporting satisfactory sleep quality (OR = 3.08; p = 0.0001). The multivariable model demonstrated good fit (Likelihood Ratio Test, p < 0.001), and the proportional odds assumption was satisfied (p = 0.46). Conclusions: These findings underscore the important role of psychological distress and sleep health in EE behavior, independent of other sociodemographic, academic, lifestyle, and anthropometric factors. Given the observed associations, interventions incorporating stress-management and sleep-improvement components may represent promising approaches for addressing emotional eating among university students; however, longitudinal and intervention studies are needed to establish causality. Future longitudinal and intervention studies are needed to establish causality among perceived stress, sleep quality, and emotional eating. Full article

Water scarcity poses significant challenges to urban water management for long-term sustainability in hyper-arid regions. This study presents a longitudinal assessment of water demand and supply dynamics in Sharjah, United Arab Emirates, from 2016 to 2022 using operational data obtained from the Sharjah Electricity, Water and Gas Authority (SEWA). Temporal trends in water production, sectoral consumption, and source transitions were evaluated using descriptive statistics and operational performance metrics: Demand–Production Ratio (DPR), Operational Production Margin (OPM), Production Adequacy Index (PAI), Source Dependency Ratio (SDR), and Seasonal Variability Index (SVI). Results showed that Sharjah maintained a production capacity consistently exceeding billed consumption, with an average DPR of approximately 72% and a PAI of 1.0. Desalinated water became the dominant supply source (~90%), while groundwater reliance declined substantially to support aquifer conservation. The residential sector accounted for approximately 62% of total water demand, highlighting the importance of demand-side management strategies. Seasonal variability analysis indicated peak demand during summer months, while desalination capacity supported relatively stable supply conditions throughout the year. The findings provide localized empirical insights into urban water management, supply diversification, and long-term water security in a hyper-arid Gulf city. Full article

Background: Normobaric hypoxia is increasingly used as an additional environmental stimulus in training programs; however, its practical role in improving anaerobic performance remains unclear. This study analyzed how a four-week high-intensity interval training (HIIT) program performed under different hypoxic exposure models affected selected anaerobic performance indices in young men. Methods: Forty-five healthy, non-elite physically active men completed the study. Participants were assigned to one of four groups: control, normoxic training, live low–train high group, and live high–train low group. The training intervention consisted of 12 HIIT sessions performed over four weeks. Anaerobic performance was assessed before and after the intervention using a 20-s maximal cycling sprint test. Results: No significant group × time interactions were found for any variables obtained from the 20-s maximal cycling sprint test, including peak power (PP; p = 0.109, ηp² = 0.136), relative peak power (rel_PP; p = 0.641, ηp² = 0.040), relative mean power (rel_MP; p = 0.438, ηp² = 0.063), relative total work (rel_TW; p = 0.515, ηp² = 0.054), fatigue index (FI; p = 0.231, ηp² = 0.099), and time to peak power (tPP; p = 0.673, ηp² = 0.036). Conclusions: The findings suggest that, under the applied conditions, normobaric hypoxia does not provide a sufficient additional stimulus to enhance anaerobic performance adaptations. From an applied exercise physiology perspective, the specificity and structure of the training stimulus are critical when designing short-term HIIT interventions. They may be more important than hypoxic exposure itself for improving short-duration maximal cycling sprint performance. Full article

The sustainable management of land resources and the formulation of land policies are closely linked to the stability and health of terrestrial ecological systems, which in turn underpin sustainable regional economic, social, and environmental development. However, land use change has a time effect on the environment and requires long-term observation to discover its impact on landscape patterns. The Yellow River Basin functions as a critical ecological barrier in northern China, where land use changes are particularly intense in the transitional zone between its middle and lower reaches. Using Landsat imagery as the data source, this study adopts the Random Forest (RF) algorithm to classify eight sets of sequential data covering a 35-year period from 1990 to 2025 in the study area. Landscape pattern metrics and transfer matrices are employed to conduct qualitative and quantitative analyses of the spatiotemporal dynamics of land use changes. Additionally, land expansion analysis strategies and the RF algorithm are applied to identify the relative importance of different driving factors. The results show that: (1) The classification accuracy based on the Google Earth Engine (GEE) cloud platform remains consistently high, exceeding 90% across all phases. (2) Patch density decreases significantly, while the largest patch index continues to decline; the Shannon diversity index shows a fluctuating upward trend, and the aggregation index exhibits a slight increase. (3) Mutual conversions among farmland, forest, and grassland are the dominant processes driving land use changes in the region. (4) The Digital Elevation Model (DEM), construction land area distribution, and distance to primary roads are the key factors influencing land use patterns, with human activities acting as the primary driver of land use type transformations in the area. Full article

Highway–rail grade crossing (HRGC) incidents in the United States declined substantially for several decades before stabilizing in recent years. Understanding this persistence is important because future safety improvements may depend on identifying locations where incident occurrence remains resistant to further reduction. This study developed an integrated framework to characterize persistent HRGC incident environments using 50 years (1976–2025) of Federal Railroad Administration incident records. Trend, structural-break, variance, and stationarity tests were first applied to determine whether the historical decline transitioned into a distinct persistence regime. A county-level persistence index (PI) was then developed to quantify the combined effects of incident burden and resistance to decline during the plateau period. Distributional analysis characterized the statistical behavior of the PI, while global and local Moran’s I statistics evaluated its spatial organization. Explainable machine learning methods were subsequently used to identify incident characteristics associated with elevated persistence. The results identified a statistically significant regime change around 2010. Prior to 2010, incidents exhibited a strong declining trend, whereas the subsequent period displayed a statistically significant but substantially weaker decline, lower variance, and behavior consistent with a persistence regime characterized by a markedly attenuated rate of improvement. The PI followed a strongly right-skewed distribution that was best represented by a bounded heavy-tailed unit log-logistic model, indicating that persistence is concentrated within a relatively small subset of counties. Spatial analysis revealed significant positive spatial autocorrelation (Moran’s I = 0.180, p = 0.001) and geographically coherent clusters concentrated primarily in the southeastern United States and several major freight-oriented regions. Explainable machine learning models identified train-operating characteristics, warning device contexts, movement patterns, and temporal conditions as key attributes associated with high-persistence counties. The findings demonstrate that the post-2010 incident plateau is sustained disproportionately by a limited number of geographically concentrated environments and provide a framework for supporting more targeted safety interventions. Full article

Sheep hemoglobin hydrolysate (SHH) has important biological antioxidant functions. However, the specific mechanism is not well understood. The aim of this study was to investigate the antioxidant activity of SHH in vivo and its potential mechanism. Eighteen Sprague Dawley rats were randomly divided into DG (gavaged with 0.9% saline), VC (gavaged with vitamin C 100 mg/kg BW) and SHH (gavaged with SHH 800 mg/kg BW) groups. The rats were given intragastric administration for 21 days. On the 22nd day, all rats were intraperitoneally injected with diquat as a challenge. The results showed that SHH treatment had no obvious effect on the growth performance and main organ index of rats after the diquat challenge. Compared to the DG group, the SHH group maintained higher serum T-AOC and GSH-Px levels after the diquat challenge. SHH significantly increased GSH-Px activity in the liver and kidney, and decreased MDA content in the kidney after the diquat challenge. SHH significantly up-regulated the gene expression of TrxR in the liver and Nrf2 in the kidney but had no significant effect on the expression of NF-κB in the kidney. Histopathological observations showed that SHH could improve the morphology of liver cells, and no obvious pathological damage was found in all organs. In conclusion, the relative decrease in oxidative stress markers observed with prophylactic SHH supplementation after diquat challenge may be mediated by the regulation of transcriptional expression of Nrf2-related redox genes. These findings demonstrate SHH’s potential as a natural antioxidant and provide promising strategies for converting low-value meat industry byproducts into functional supplements. Full article

Leaf nitrogen content (LNC) is an important physiological indicator for evaluating citrus nutritional status, photosynthetic capacity, and fertilization demand. However, conventional LNC determination mainly relies on field sampling and laboratory chemical analysis, which are destructive, time-consuming, labor-intensive, and limited in spatial continuity, making them unsuitable for large-scale real-time nitrogen monitoring in complex orchard environments. To achieve rapid and non-destructive estimation of citrus LNC, this study developed a UAV multispectral inversion framework integrating object-based canopy extraction and machine learning models. Field experiments were conducted in a citrus orchard in western Hubei Province, China. Multi-temporal UAV multispectral images were collected from April to October 2025, and ground measurements of citrus LNC were collected simultaneously. First, minimum distance classification (MDC), maximum likelihood classification (MLC), and object-based image analysis (OBIA) were used for land-cover classification of citrus orchard images, and their canopy extraction performance under complex orchard backgrounds was compared. Subsequently, multiple vegetation indices were calculated from the extracted citrus canopy spectra, and sensitive spectral features were selected through correlation analysis. Finally, seven models, including simple linear regression, quadratic regression, partial least squares regression (PLS), back propagation neural network (BP), extreme learning machine (ELM), particle swarm optimization-extreme learning machine (PSO-ELM), and particle swarm optimization-back propagation neural network (PSO-BP), were constructed to systematically evaluate the inversion performance of citrus LNC across the entire growth period. The results showed that: (1) OBIA achieved higher classification accuracy and temporal stability in citrus orchard land-cover classification, with overall accuracy ranging from 68.86% to 85.65% and Kappa coefficients ranging from 0.56 to 0.72, outperforming MDC and MLC. This indicates that OBIA can effectively reduce the interference of bare soil, grass, shadows, and other non-target objects on canopy spectral extraction. (2) The correlations between vegetation indices and LNC varied markedly among different growth stages, suggesting that the spectral response of citrus LNC has strong phenological dependence and that a single vegetation index is insufficient to stably characterize LNC variation across the whole growth period. (3) At the whole-growth-period scale, multi-index fusion models outperformed single-index models, among which EVI, TVI, and MTVI showed relatively strong cross-stage sensitivity. (4) Optimized machine learning models generally outperformed traditional regression models and unoptimized machine learning models. Among them, PSO-BP achieved the best performance, with a validation R² of 0.68 and an RMSE of 1.54 g kg⁻¹, representing an increase in R² of 23.64% compared with the PLS model and 25.93% over the baseline BP model in terms of R². Overall, this study demonstrates that OBIA-based canopy spectral quality improvement combined with PSO-optimized machine learning can effectively improve the stability and reliability of UAV multispectral estimation of citrus LNC under complex orchard backgrounds. The proposed framework provides technical support for citrus nitrogen diagnosis, precision fertilization, and intelligent orchard management. Full article