Search Results (12,653)

Polarimetric synthetic aperture radar (PolSAR) calibration quality assessment is essential for verifying the reliability of polarimetric calibration results and ensuring the accuracy of subsequent quantitative applications. The corner-reflector-based assessment is accurate but depends on field deployment and maintenance, whereas the distributed-target-based methods is easier to automate but is sensitive to mixed scattering within image patches and to non-unique histogram peaks. In addition, the parameter distribution may contain multiple peaks, affecting the uniqueness and stability of assessment. To address these problems, this paper proposes a robust distributed-target-based PolSAR calibration quality assessment (RD-PCQA) method without corner reflectors (CRs) to address these problems. The proposed method first uses hypothesis testing of confidence interval method for PolSAR calibration (PCHTCI) to extract high-quality distributed targets and combines the polarimetric correlation coefficient R_HHVV to select volume-scattering-dominant targets, thereby improving the physical consistency of samples used for channel imbalance amplitude (CIA) estimation. Second, a high-proportion distributed-target constraint is used to refine the samples for channel imbalance phase (CIP) and polarimetric crosstalk estimation, reducing the influence of nonideal scatterers on parameter estimation. Finally, a unique peak searching strategy based on progressively enlarged statistical scales is proposed to suppress the effect of multi-peak distributions on assessment result. Experiments were conducted using three GF-3 PolSAR images acquired over the SAR calibration site in Etuoke Banner, Ordos, Inner Mongolia, China, with CR results used as references, and considering finite sample uncertainty. The experimental results show that, compared with the conventional distributed-target-based method, the proposed method is closer to the corresponding CR mean in all comparisons, with the mean absolute deviations for CIA, CIP and crosstalk scenarios reduced to 0.024 dB, 2.043° and 3.069 dB, respectively. Therefore, it demonstrates the effectiveness and practical potential of the proposed method for PolSAR calibration quality assessment without CRs. Full article

Hyperspectral image (HSI) classification benefits from rich spectral information; however, high dimensionality of HSI data increases computational cost, noise sensitivity, and the risk of overfitting when labeled samples are limited. Most pretrained computer vision networks are designed for three-channel inputs, making direct application to hyperspectral cubes difficult. Conventional principal component analysis (PCA) ranks components by total variance without distinguishing useful signal variance from noise-related variance, which can reduce the reliability of the resulting representation when only a few components are retained. This paper proposes a data-augmented Noise-Adjusted Principal Component Analysis (DA-NAPCA) framework for deep learning-based HSI classification. By accounting for estimated noise covariance, NAPCA orders the transformed components by signal-to-noise ratio rather than total variance, while data augmentation mitigates the overfitting risk when labeled samples are limited. Unlike typical NAPCA/MNF applications, which select the number of retained components empirically, DA-NAPCA deliberately retains three noise-adjusted components to form a compact three-channel representation, enabling pretrained models designed for three-channel inputs to be fine-tuned without modifying their input layers. The framework is evaluated using a 3D convolutional neural network (3D-CNN) for spatial–spectral feature learning and a pretrained EfficientNet-B0 model for lightweight transfer learning. Although this paper uses 3D-CNN and EfficientNet-B0 as illustrative examples, the proposed DA-NAPCA framework is a representation-level preprocessing approach and does not require architecture-specific modification. Experiments conducted on the Indian Pines, University of Pavia, and Salinas datasets compare DA-NAPCA with RGB, band selection, PCA-based dimensionality reduction, and ablation variants. Across the three datasets, DA-NAPCA achieved mean overall accuracies of 93.11–94.71% with 3D-CNN and 95.93–97.44% with EfficientNet-B0. Compared with the second-best baseline method, DA-NAPCA improved overall accuracy by 2.75–7.58 percentage points with 3D-CNN and 1.28–2.12 percentage points with EfficientNet-B0. These results demonstrate that combining a compact noise-adjusted representation with spatial augmentation provides an effective input representation for deep learning-based HSI classification. Full article

Text-to-image generation has rapidly advanced the creation of digital artwork, yet most existing models rely on centralized training pipelines that require collecting large-scale image–text pairs from artists, design studios, or online communities. Such centralized practice raises serious privacy, ownership, and style leakage concerns, especially when local datasets contain identifiable artistic signatures or proprietary visual assets. To address this problem, this paper proposes FedMuse, a privacy-aware federated framework for multi-style text-to-image art generation, in which distributed clients collaboratively train a shared generative model while keeping their private art data local. The proposed framework decomposes the learning process into three coordinated components: a global semantic alignment module that captures cross-client text–image correspondence, a local style adapter that preserves client-specific artistic characteristics, and a privacy-calibrated aggregation mechanism that suppresses sensitive style leakage during model update exchange. To further improve multi-style generation, we design a style-disentangled federated optimization algorithm that separates content-relevant knowledge from client-private stylistic representations, allowing the global model to generalize across diverse artistic domains without directly absorbing private local styles. In addition, an adaptive privacy regularizer is introduced to reduce memorization risk while maintaining visual quality and prompt consistency. Experiments on real-world text-to-image art datasets demonstrate competitive generation quality, stronger personalization, and improved empirical resistance to membership and style-leakage attacks. Because the calibration statistics are data-dependent, FedMuse does not claim a certified end-to-end differential-privacy budget. The results suggest that privacy-aware collaboration can be a practical direction for distributed AI art generation. Full article

With the continuous improvement in the high-speed railway (HSR) network, passengers’ holiday travel demand and the scale of tourism passenger flow have grown rapidly, placing higher requirements on HSR transportation services. Current train line plans are formulated according to regular daily passenger flow and fail to adapt to the travel demand generated by various holiday tourism products, resulting in poor adaptation to holiday passenger flow characteristics. Different from existing line-planning studies that consider only regular daily passenger flow, this paper is among the first to embed hierarchical holiday tourism products as a structural input of the HSR line planning problem and to explicitly handle asymmetric passenger demand in holiday periods. To address this gap, this paper optimizes HSR line plans based on the travel characteristics of tourism products. An integer programming model is established that allocates asymmetric passenger flows to train flows while minimizing the total operating cost. The model incorporates both conventional HSR passenger demand and the differentiated travel demand corresponding to different tourism products. A real-world experiment based on the HSR network in Jiangxi Province, China demonstrates that the proposed model effectively improves the tourism transportation efficiency and passenger flow distribution of HSR with reliable practicability. Sensitivity analyses further reveal how the tourism demand scale, the tourism sub-product diversity, and the benchmark line requirements affect the feasibility and cost of the line plan. This study provides a valid reference for the optimization of holiday train line plans and the coordinated development of tourism and transportation. Full article

High-pressure premixed H₂/O₂ combustion in closed vessels involves short chemical time scales, strong compression-wave feedback, and repeated end-wall reflections, making the onset of deflagration-to-detonation transition (DDT) highly sensitive to gas-dynamic–chemical coupling. This review focuses on DDT developing from the acceleration of an initially premixed flame in closed, high-pressure H₂/O₂ systems. Studies of H₂/air, diluted mixtures, and open or semi-closed configurations are considered only where they provide relevant mechanistic or methodological insight, and their applicability to closed, high-pressure, undiluted H₂/O₂ conditions is assessed explicitly. The review synthesizes confined flame acceleration, induction-time gradients, coherent energy release, and shock–flame interaction together with H₂/O₂ chemical kinetics, compressible reactive-flow modeling, multidimensional simulations, and model validation. The reviewed evidence indicates that the transition is governed by the coupled evolution of flame-area growth and compression waves, restructuring of the induction-time field, local energy release, and the establishment of persistent shock–reaction coupling. Initial pressure and temperature, mixture distribution, ignition strategy, geometry, vessel scale, and thermal boundaries influence this process by modifying intrinsic reaction scales, wave phasing, reflection paths, and heat and momentum losses. Numerical predictions remain sensitive to chemical kinetics, transport treatment, numerical dissipation, wall modeling, dimensionality, mesh resolution, and the definition of the DDT event. Accordingly, DDT identification based only on peak pressure or a short interval of near-CJ wave velocity can be ambiguous, whereas multi-observable validation using pressure, wave or flame position, reaction-zone information, and coupling persistence provides a more reliable basis for model assessment. The principal remaining gap is the limited availability of complete-process, three-dimensional experiment–simulation datasets for fully closed vessels containing high-pressure, undiluted H₂/O₂. Full article

Classroom behavior detection plays an important role in intelligent education by providing objective visual evidence for the analysis of learning engagement, classroom interaction assessment, and teaching evaluation. However, accurate behavior detection for individual students in real classroom environments remains challenging due to subtle behavior-related cues, visually similar action categories, occlusion, and complex background interference. To address these challenges, this paper proposes a Fine-grained Gated Detail Network (FGD-Net) for the fine-grained detection of individual student behaviors in classroom scenes. The proposed network improves behavior representation from three complementary aspects. Firstly, a Fine-grained Dynamic Recalibration Convolution Block (FDRC) is designed to enhance behavior-sensitive local regions, such as hands, arms, heads, and upper-body postures. Secondly, a Multi-path Gated Context Aggregation Block (MGCA) is introduced to aggregate complementary contextual information from multiple feature paths, thereby strengthening the semantic representation of visually similar classroom behaviors. Thirdly, a Shift-guided Detail Reconstruction Upsampling Block (SDRU) is developed to alleviate spatial detail loss during multi-scale feature fusion and improve the reconstruction of small-scale behavior-related cues. Extensive experiments are conducted on the SCB5 and SCB3 classroom behavior detection datasets. The experimental results show that FGD-Net achieves better detection performance than recent detectors on both datasets. These results demonstrate the potential of FGD-Net as an efficient visual perception model for scalable classroom video analytics and intelligent education applications. Full article

Rapid and non-destructive monitoring of soil total nitrogen (STN) is important for precision nutrient management in ecologically fragile agricultural systems. This study established a controlled spectral resampling experiment to simulate the multispectral responses of Sentinel-2, WorldView-3, GF-6, and Landsat-9 from laboratory ASD hyperspectral measurements of Hemerocallis citrina fields. Two-dimensional (DI, RI, and NDI) and three-dimensional (TBI1–TBI5) spectral indices were constructed and evaluated using eight machine learning algorithms across five phenological stages. The results demonstrate that: (1) Three-dimensional spectral indices exhibited substantially higher sensitivity to STN than conventional two-dimensional indices, with TBI3 showing the strongest overall correlation; among the simulated sensor configurations, GF-6 delivered the best mean performance due to its dual red-edge bands. (2) Genetic algorithm-optimized backpropagation neural network (GA-BPNN) effectively addressed the local-minima limitation of standard backpropagation neural networks (BPNNs) and displayed robust generalization under multi-sensor and multi-phenological scenarios. (3) Phenology-specific modeling reduced spectral heterogeneity caused by pooling across growth stages, increasing R² by 20.28% and decreasing RMSE by 18.35%, with leaf expansion and bolting identified as optimal estimation windows. These findings elucidate the spectral response potential explanations of STN across phenological stages and provide a reference for applying multi-source simulated remote sensing data to nutrient monitoring in specialty agricultural systems. Full article

Background: Postoperative peritoneal adhesions arise from a dysregulated wound-healing response in which macrophages may have context-dependent effects. We investigated the contribution of macrophage-rich peritoneal and mesenteric compartments, the origin of macrophage-like cells in ischemic lesions, and the association between CCR2-dependent recruitment and postoperative inflammatory and reparative gene expression. Methods: Using a murine ischemic-button model, we assessed the effects of clodronate liposome treatment, bone marrow chimerism, and global CCR2 deficiency. Adhesion formation, F4/80+ cell accumulation, donor-marker expression, and selected inflammatory and wound-healing-associated transcripts were analyzed at predefined postoperative time points. Results: Clodronate liposome treatment was associated with reduced adhesion formation and substantial depletion of F4/80+ cells in peritoneal lavage and mesenteric tissue. However, F4/80+ cell numbers within ischemic buttons at postoperative day 3 were not significantly reduced, indicating that the depletion experiment does not establish selective depletion of all lesional macrophages. Bone marrow chimera experiments identified donor-marker-positive, F4/80+ cells within ischemic buttons, supporting recruitment of hematopoietic cells with a macrophage-like phenotype. CCR2 deficiency reduced F4/80+ cell accumulation in ischemic buttons and was associated with increased adhesion scores and altered expression of inflammatory and wound-healing-associated genes. These findings identify differential associations of clodronate-sensitive macrophage-rich compartments and CCR2-dependent hematopoietic recruitment with postoperative adhesion formation. Conclusions: Depletion of macrophage-rich peritoneal and mesenteric compartments was associated with reduced adhesion formation, whereas global CCR2 deficiency was associated with fewer lesional F4/80+ cells and greater adhesion severity. Because clodronate depletion, F4/80 staining, bone marrow chimerism, and global CCR2 deficiency do not provide cell-specific or fate-mapped resolution, these data do not establish distinct resident versus infiltrating macrophage functions or a reparative phenotype of CCR2-dependent cells. Cell-specific and temporally resolved validation is required to define the contributions and temporal relationships of individual macrophage subsets. Full article

Background/Objectives: Oral squamous cell carcinoma (OSCC) has high mortality rates and leads to serious health problems when diagnosed late. This situation is considered a public health problem. Histopathological examination, which is an important point in the diagnosis of the disease, is a time-consuming and manual process that requires expertise. Methods: This study presents a deep learning-based approach for the automated classification of normal oral epithelium and oral squamous cell carcinoma (OSCC) from histopathological images. The performance of state-of-the-art architectures such as Vision Transformer (ViT), CLIP, ConvNextV2, Deit, and Dinov2 was comparatively analyzed. Based on the results, a hybrid architecture combining the strengths of the models is proposed. Results: In experiments conducted on a dataset of 696 histopathological images, the ConvNextV2 and ViTL16 architectures stood out among the basic models with accuracy rates around 84%. However, the most significant contribution of this study is the proposed method, which combines the global context capability of Transformer-based models with the local feature extraction power of CNN-based models using the Efficient Channel Attention (ECA)—Gated Features model. This hybrid model, created by integrating the ViTL16, Dinov2, and ConvNextV2 architectures, achieved 89.95% accuracy, 89.82% F1 score, and 89.95% sensitivity with a KNN classifier, outperforming the baseline models in the literature. Conclusions: The results obtained demonstrate that the fusion of multiple architectures increases diagnostic reliability in medical image analysis and can assist pathologists as a decision support mechanism. Full article

Background/Objectives: Family caregivers of people with schizophrenia experience substantial psychological and practical demands associated with long-term caregiving. This study examined whether participation in a web-based digital psychoeducation program was associated with longitudinal changes in coping mechanisms among family caregivers of people with schizophrenia. Methods: A non-randomized quasi-experimental pretest–posttest control-group study was conducted at a mental health hospital in Indonesia. The final analytic sample comprised 202 family caregivers of people with schizophrenia, with 101 participants in the intervention group and 101 in the control group. The intervention comprised 10 structured web-based psychoeducation modules delivered over 10 weeks. Coping was assessed using the Brief COPE Inventory and summarized into problem-focused, emotion-focused, and avoidance coping domains. Linear mixed-effects models were used as the primary adjusted longitudinal analysis, with study group, time, group-by-time interaction, age group, and marital status included as fixed effects. Wilcoxon signed-rank and Mann–Whitney U tests were retained as complementary unadjusted analyses, while baseline-adjusted linear regression models were used as sensitivity analyses. Results: Significant group-by-time interactions were observed for problem-focused coping and emotion-focused coping. The adjusted between-group difference in change was 0.450 points (95% CI: 0.357–0.544; partial ηp² = 0.312; p < 0.001) for problem-focused coping and 0.263 points (95% CI: 0.137–0.390; partial ηp² = 0.078; p < 0.001) for emotion-focused coping. In contrast, no statistically significant differential change was observed for avoidance coping (adjusted difference in change = 0.074 points, 95% CI: −0.055 to 0.204; partial ηp² = 0.006; p = 0.259). Baseline-adjusted sensitivity analyses produced a consistent overall pattern. Conclusions: Participation in the 10-week web-based digital psychoeducation program was associated with greater longitudinal improvements in problem-focused and emotion-focused coping, but not avoidance coping. Given the non-randomized design and baseline differences between groups, these findings should be interpreted as adjusted longitudinal associations rather than definitive evidence of causal efficacy. The program may therefore represent an accessible complementary approach to supporting caregiver adaptation in long-term schizophrenia care. Full article

Background: Smartphone-based store-and-forward teledermatology (SAF-TD) has emerged as a potential approach to improving access to dermatological care. Patient-initiated SAF-TD systems, which enable direct image submission to dermatology specialists, represent an increasingly relevant but insufficiently studied model. However, diagnostic agreement remains insufficiently characterized in real-world, patient-initiated, direct-to-specialist systems in which adult patients independently submit conventional smartphone photographs and assessments are performed by dermatologists with different levels of clinical experience. Methods: A prospective cross-sectional, single-center study was conducted between February 2022 and September 2024, which included 83 patients. Patients independently submitted conventional smartphone photographs through a hospital patient portal. Diagnoses established remotely by an experienced and a beginner dermatologist were compared with face-to-face (FTF) diagnoses as the reference standard. Melanocytic lesions were defined as the positive category and non-melanocytic conditions as the negative category. Diagnostic agreement was assessed using Cohen’s kappa, and sensitivity and specificity were calculated with 95% confidence intervals (CIs). Results: Agreement between SAF-TD and FTF classification was 91.6% (95% CI, 83.6–95.9) for the experienced dermatologist (κ = 0.781; 95% CI, 0.628–0.934) and 90.4% (95% CI, 82.1–95.0) for the beginner dermatologist (κ = 0.760; 95% CI, 0.603–0.917), indicating substantial agreement. Sensitivity and specificity were 94.7% (95% CI, 75.4–99.1) and 90.6% (95% CI, 81.0–95.6), respectively, for the experienced dermatologist and 86.4% (95% CI, 66.7–95.3) and 91.8% (95% CI, 82.2–96.4), respectively, for the beginner dermatologist. Conclusions: Patient-initiated smartphone-based SAF-TD demonstrated substantial agreement with FTF binary classification by both dermatologists. These findings support its potential clinical use as a complementary method for initial remote dermatological assessment. However, the results require validation in larger, multicenter studies. Full article

To address the resource utilization of waste tires and the fire-safety concerns in engineering applications of rubber concrete (RC), this study systematically investigates the residual static and dynamic mechanical properties of RC after exposure to elevated temperatures and subsequent cooling to room temperature. Specimens are prepared by replacing fine aggregate with rubber particles at equal volume replacement ratios of 0%, 5%, 15%, and 30%. After undergoing gradient heating to target temperatures ranging from 20 °C to 300 °C, the specimens are naturally cooled to room temperature prior to testing. Subsequently, static compressive and splitting tensile tests, along with dynamic impact tests using a Split Hopkinson Pressure Bar (SHPB), are performed. These experiments are supplemented by scanning electron microscopy (SEM) to elucidate the microscale mechanisms. The results show that the residual static strength decreases monotonically with increasing rubber content and temperature. For the 30% rubber content mixture, the compressive strength decreased by approximately 40.6% from ambient temperature to 300 °C, and its strength is 64.6% lower than that of NC at 300 °C. Dynamic strength exhibits a pronounced strain-rate effect, with the strain-rate sensitivity of DIF being enhanced by higher rubber content. Energy dissipation increases substantially with strain rate; rubberized mixtures generally exhibit higher energy dissipation than NC at lower strain rates, though this effect becomes less evident at higher strain rates. These findings provide a theoretical foundation for the application of RC in complex thermo-mechanical loading scenarios, particularly in evaluating its post-fire residual load-bearing capacity. Full article

Emotional information often modulates recognition memory, but the relative roles of arousal and valence remain debated, partly because many studies rely on normative emotion categories rather than individualized affective experience. This study examined whether post-test individualized arousal and valence ratings were associated with recognition-confidence responses to emotional Chinese words and whether the learning task influenced later recognition. Forty participants studied neutral, positive, and negative words under semantic-judgment and recognition-judgment learning conditions. After a 24 h delay, they completed a six-point old/new confidence test and then rated each final-test word for valence and arousal. Linear mixed-effects models showed that individualized arousal was more consistently associated with stronger old-response confidence than individualized valence, with a nonlinear increase at higher arousal levels. Semantic-judgment learning was followed by higher old-item confidence and hit probability than recognition-judgment learning. Descriptive signal-detection summaries indicated that valence-category effects were more evident in false-alarm rates, response criterion, and d′ (sensitivity index) estimates than in the primary old-response confidence model. Because affective ratings were collected after recognition, these findings should be interpreted as associative rather than causal, and they highlight the need to distinguish recognition confidence, response bias, and memory sensitivity in emotional memory research. Full article

Background/Objectives: Morphological signs of non-carious tooth surface loss are clinically heterogeneous, and their prevalence depends strongly on the diagnostic framework and threshold used. This exploratory secondary analysis evaluated a study-specific combined morphological outcome restricted to second permanent molars and its association with cumulative caries experience in a regional clinical sample of adolescents. Methods: The analysis included 231 adolescents aged 13–19 years from Dolj, Gorj, and Olt counties, selected from a parent clinical database of 638 participants. The study-specific outcome was coded when a clearly discernible non-carious hard-tissue change compatible with an abrasive, erosive, or abfraction-related morphology was identified on any clinically accessible surface of tooth 17, 27, 37, or 47. This study did not use BEWE, the Smith and Knight Tooth Wear Index, or another validated severity-based tooth wear index; subtype, surface, and severity were not retained separately in the analytical dataset. Dental caries experience was measured using DMF-T. Group comparisons, correlations, and multivariable logistic regression were performed in JASP version 0.96.0. Sensitivity analyses included a participant-level coding audit and a logistic model using DMF-T recalculated after excluding the contribution of teeth 17, 27, 37, and 47. Results: The study-specific morphological outcome was present in 50/231 participants (21.6%). A coding audit confirmed that 181 participants had no positive second molar and 50 had exactly one positive second molar; tooth-specific counts were 14, 15, 12, and 9 for teeth 17, 27, 37, and 47, respectively. In the primary model, male sex (adjusted OR = 3.34, 95% CI 1.67–6.71) and DMF-T (adjusted OR = 1.15 per point, 95% CI 1.01–1.31) were associated with the outcome. After excluding the four evaluated second molars from DMF-T, the caries experience association remained positive but borderline (OR = 1.198, 95% CI 1.000–1.435, p = 0.049), while male sex remained associated (OR = 3.310, 95% CI 1.647–6.652, p < 0.001). Conclusions: A modest exploratory association was observed between the study-specific morphological outcome and cumulative caries experience in this clinical sample. The result should not be interpreted as a standardized NCDL prevalence estimate, a causal relationship, or a prediction model. Confirmation using validated whole-mouth tooth wear measures, outcome-specific reliability assessment, and population-based longitudinal data is required. Full article

Background/Objectives: Accurate delineation of the optic disc and optic cup in retinal fundus photographs is a prerequisite for automated glaucoma screening. While encoder–decoder segmentation models have advanced considerably, the contribution of upstream preprocessing to segmentation accuracy, and the stability of that contribution across repeated training runs, remain insufficiently characterized. Methods: Five preprocessing pipelines, baseline, Contrast Limited Adaptive Histogram Equalization (CLAHE), Region of Interest (ROI) cropping, ROI+CLAHE, and CLAHE with heavy augmentation, were benchmarked under a fixed EfficientUNet++ model with an EfficientNet-B7 encoder on three publicly available fundus datasets (REFUGE, ORIGA, and Drishti-GS). Every configuration was retrained under three independent random seeds (42, 15, and 89) to assess run-to-run variability. Seed-level standard deviations accompany every reported mean and define the confidence limit on each ranking. Results: On REFUGE, CLAHE with augmentation (Config 5) achieved the strongest mean Dice (disc $0.9523\pm 0.0017$; cup $0.8348\pm 0.0018$). On ORIGA, all five configurations clustered within 0.0067 disc Dice; ROI+CLAHE (Config 4) was marginally ahead on disc ($0.9681\pm 0.0002$) and augmentation led on the cup ($0.8873\pm 0.0024$). On Drishti-GS, all five configurations converged successfully once optimizer and loss settings were corrected; the near-total failures seen in earlier single-run experiments reflected a configuration problem, not the small (81-image) training set. Conclusions: CLAHE applied to full-resolution images is the single most consistently beneficial preprocessing choice across all three datasets. ROI+CLAHE showed a small, initialization-stable advantage on ORIGA, but ROI crop centres were derived from ground-truth centroids, an oracle localization setting, and these results should not be interpreted as achievable by a fully automated pipeline. Data augmentation showed a consistent reduction in initialization sensitivity on small datasets and may be beneficial as a default strategy. Full article