Search Results (100)

Understanding the intricate relationship between sensory perception and physicochemical properties of cacao-based products is crucial for advancing quality control and driving product innovation. However, effectively integrating these heterogeneous data sources poses a significant challenge, particularly when sensory evaluations are derived from low-quality, subjective, and often inconsistent annotations provided by multiple experts. We propose a comprehensive framework that leverages a correlated chained Gaussian processes model for learning from crowds, termed MAR-CCGP, specifically designed for a customized Casa Luker database that integrates sensory and physicochemical data on cacao-based products. By formulating sensory evaluations as regression tasks, our approach enables the estimation of continuous perceptual scores from physicochemical inputs, while concurrently inferring the latent, input-dependent reliability of each annotator. To address the inherent noise, subjectivity, and non-stationarity in expert-generated sensory data, we introduce a three-stage methodology: (i) construction of an integrated database that unifies physicochemical parameters with corresponding sensory descriptors; (ii) application of a MAR-CCGP model to infer the underlying ground truth from noisy, crowd-sourced, and non-stationary sensory annotations; and (iii) development of a novel localized expert trustworthiness approach, also based on MAR-CCGP, which dynamically adjusts for variations in annotator consistency across the input space. Our approach provides a robust, interpretable, and scalable solution for learning from heterogeneous and noisy sensory data, establishing a principled foundation for advancing data-driven sensory analysis and product optimization in the food science domain. We validate the effectiveness of our method through a series of experiments on both semi-synthetic data and a novel real-world dataset developed in collaboration with Casa Luker, which integrates sensory evaluations with detailed physicochemical profiles of cacao-based products. Compared to state-of-the-art learning-from-crowds baselines, our framework consistently achieves superior predictive performance and more precise annotator reliability estimation, demonstrating its efficacy in multi-annotator regression settings. Of note, our unique combination of a novel database, robust noisy-data regression, and input-dependent trust scoring sets MAR-CCGP apart from existing approaches. Full article

Electricity demand forecasting is essential for smart grid management, yet it presents challenges due to the dynamic nature of consumption trends and regional variability in usage patterns. While federated learning (FL) offers a privacy-preserving solution for handling sensitive, region-specific data, traditional FL approaches struggle when local datasets are limited, often leading models to overfit noisy peak fluctuations. Additionally, many regions exhibit stable, periodic consumption behaviors, further complicating the need for a global model that can effectively capture diverse patterns without overfitting. To address these issues, we propose Federated Hybrid Graph Attention Network with Two-step Optimization for Electricity Consumption Forecasting (FedHMGAT), a hybrid modeling framework designed to balance periodic trends and numerical variations. Specifically, FedHMGAT leverages a numerical structure graph with a Gaussian encoder to model peak fluctuations as dynamic covariance features, mitigating noise-driven overfitting, while a multi-scale attention mechanism captures periodic consumption patterns through hybrid feature representation. These feature components are then fused to produce robust predictions. To enhance global model aggregation, FedHMGAT employs a two-step parameter aggregation strategy: first, a regularization term ensures parameter similarity across local models during training, and second, adaptive dynamic fusion at the server tailors aggregation weights to regional data characteristics, preventing feature dilution. Experimental results verify that FedHMGAT outperforms conventional FL methods, offering a scalable and privacy-aware solution for electricity demand forecasting. Full article

This study proposes a geographically weighted (GW) quantile machine learning (GWQML) framework for soil moisture (SM) prediction, integrating spatial kernel functions with quantile-based prediction and uncertainty quantification. The framework incorporates satellite radar backscatter, meteorological re-analysis, and topographic variables, applied across 15 SM stations and six land use systems at the North Wyke Farm Platform, southwest England, UK. GWQML was implemented using Gaussian and Tricube spatial kernels across a range of kernel bandwidths (500–1500 m). Model performance was evaluated using both in-sample and Leave-One-Land-Use-Out validation schemes, and a global quantile machine learning model (QML) without spatial weighting served as the benchmark. GWQML achieved R² values up to 0.85 and prediction interval coverage probabilities up to 0.9, with intermediate kernel bandwidths (750–1250 m) offering the best balance between accuracy and uncertainty calibration. Spatial autocorrelation analysis using Moran’s I revealed a lower residual clustering under GWQML relative to the benchmark model, which suggests improved handling of local spatial variation. This study represents one of the first applications of geographically weighted kernel functions in a quantile machine learning framework for daily soil moisture prediction. The approach implicitly captures spatially varying relationships while delivering calibrated uncertainty estimates for scalable SM monitoring across heterogenous agricultural landscapes. Full article

Structural collapses caused by accidents or disasters could create unexpected radiation shielding, resulting in sharp gradients within the radiation field. Traditional radiation mapping methods often fail to accurately capture these complex variations, making the rapid and precise localization of radiation sources a significant challenge in emergency response scenarios. To address this issue, based on standard Gaussian process regression (GPR) models that primarily utilize a single Gaussian kernel to reflect the inverse-square law in free space, a novel multi-kernel Gaussian process regression (MK-GPR) model is proposed for high-fidelity radiation mapping in environments with physical obstructions. MK-GPR integrates two additional kernel functions with adaptive weighting: one models the attenuation characteristics of intervening materials, and the other captures the energy-dependent penetration behavior of radiation. To validate the model, gamma-ray distributions in complex, shielded environments were simulated using GEometry ANd Tracking 4 (Geant4). Compared with conventional methods, including linear interpolation, nearest-neighbor interpolation, and standard GPR, MK-GPR demonstrated substantial improvements in key evaluation metrics, such as MSE, RMSE, and MAE. Notably, the coefficient of determination (R²) increased to 0.937. For practical deployment, the optimized MK-GPR model was deployed to an RK-3588 edge computing platform and integrated into a mobile robot equipped with a NaI(Tl) detector. Field experiments confirmed the system’s ability to accurately map radiation fields and localize gamma sources. When combined with SLAM, the system achieved localization errors of 10 cm for single sources and 15 cm for dual sources. These results highlight the potential of the proposed approach as an effective and deployable solution for radiation source investigation in post-disaster environments. Full article

With the large-scale integration of renewable energy into distribution networks, traditional fixed-setting overcurrent protection strategies struggle to adapt to rapid fluctuations in renewable energy (e.g., wind and photovoltaic) output. Optimizing current settings is crucial for enhancing the stability of modern distribution networks. This paper proposes an adaptive overcurrent protection method based on an improved NSGA-II algorithm. By dynamically detecting renewable power fluctuations and generating adaptive solutions, the method enables the online optimization of protection parameters, effectively reducing misoperation rates, shortening operation times, and significantly improving the reliability and resilience of distribution networks. Using the rate of renewable power variation as the core criterion, renewable power changes are categorized into abrupt and gradual scenarios. Depending on the scenario, either a random solution injection strategy (DNSGA-II-A) or a Gaussian mutation strategy (DNSGA-II-B) is dynamically applied to adjust overcurrent protection settings and time delays, ensuring real-time alignment with grid conditions. Hard constraints such as sensitivity, selectivity, and misoperation rate are embedded to guarantee compliance with relay protection standards. Additionally, the convergence of the Pareto front change rate serves as the termination condition, reducing computational redundancy and avoiding local optima. Simulation tests on a 10 kV distribution network integrated with a wind farm validate the effectiveness of the proposed method. Full article

Recent advances in precision manufacturing and high-end equipment technologies have imposed ever more stringent requirements on the accuracy, real-time performance, and lightweight design of online steel strip surface defect detection systems. To reconcile the persistent trade-off between detection precision and inference efficiency in complex industrial environments, this study proposes StripSurface–YOLO, a novel real-time defect detection framework built upon YOLOv8n. The core architecture integrates an Efficient Cross-Stage Local Perception module (ResGSCSP), which synergistically combines GSConv lightweight convolutions with a one-shot aggregation strategy, thereby markedly reducing both model parameters and computational complexity. To further enhance multi-scale feature representation, this study introduces an Efficient Multi-Scale Attention (EMA) mechanism at the feature-fusion stage, enabling the network to more effectively attend to critical defect regions. Moreover, conventional nearest-neighbor upsampling is replaced by DySample, which produces deeper, high-resolution feature maps enriched with semantic content, improving both inference speed and fusion quality. To heighten sensitivity to small-scale and low-contrast defects, the model adopts Focal Loss, dynamically adjusting to sample difficulty. Extensive evaluations on the NEU-DET dataset demonstrate that StripSurface–YOLO reduces FLOPs by 11.6% and parameter count by 7.4% relative to the baseline YOLOv8n, while achieving respective improvements of 1.4%, 3.1%, 4.1%, and 3.0% in precision, recall, mAP₅₀, and mAP_50:95. Under adverse conditions—including contrast variations, brightness fluctuations, and Gaussian noise—SteelSurface-YOLO outperforms the baseline model, delivering improvements of 5.0% in mAP₅₀ and 4.7% in mAP_50:95, attesting to the model’s robust interference resistance. These findings underscore the potential of StripSurface–YOLO to meet the rigorous performance demands of real-time surface defect detection in the metal forging industry. Full article

Most image forgery localization methods rely on supervised learning, requiring large labeled datasets for training. Recently, several unsupervised approaches based on the variational autoencoder (VAE) framework have been proposed for forged pixel detection. In these approaches, the latent space is built by a simple Gaussian distribution or a Gaussian Mixture Model. Despite their success, there are still some limitations: (1) A simple Gaussian distribution assumption in the latent space constrains performance due to the diverse distribution of forged images. (2) Gaussian Mixture Models (GMMs) introduce non-convex log-sum-exp functions in the Kullback–Leibler (KL) divergence term, leading to gradient instability and convergence issues during training. (3) Estimating GMM mixing coefficients typically involves either the expectation-maximization (EM) algorithm before VAE training or a multilayer perceptron (MLP), both of which increase computational complexity. To address these limitations, we propose the Deep ViT-VAE-GMM (DVVG) framework. First, we employ Jensen’s inequality to simplify the KL divergence computation, reducing gradient instability and improving training stability. Second, we introduce convolutional neural networks (CNNs) to adaptively estimate the mixing coefficients, enabling an end-to-end architecture while significantly lowering computational costs. Experimental results on benchmark datasets demonstrate that DVVG not only enhances VAE performance but also improves efficiency in modeling complex latent distributions. Our method effectively balances performance and computational feasibility, making it a practical solution for real-world image forgery localization. Full article

Solar power generation is rapidly emerging within renewable energy due to its cost-effectiveness and ease of deployment. However, improper inspection and maintenance lead to significant damage from unnoticed solar hotspots. Even with inspections, factors like shadows, dust, and shading cause localized heat, mimicking hotspot behavior. This study emphasizes interpretability and efficiency, identifying key predictive features through feature-level and What-if Analysis. It evaluates model training and inference times to assess effectiveness in resource-limited environments, aiming to balance accuracy, generalization, and efficiency. Using Unmanned Aerial Vehicle (UAV)-acquired thermal images from five datasets, the study compares five Machine Learning (ML) models and five Deep Learning (DL) models. Explainable AI (XAI) techniques guide the analysis, with a particular focus on MPEG (Moving Picture Experts Group)-7 features for hotspot discrimination, supported by statistical validation. Medium Gaussian SVM achieved the best trade-off, with 99.3% accuracy and 18 s inference time. Feature analysis revealed blue chrominance as a strong early indicator of hotspot detection. Statistical validation across datasets confirmed the discriminative strength of MPEG-7 features. This study revisits the assumption that DL models are inherently superior, presenting an interpretable alternative for hotspot detection; highlighting the potential impact of domain mismatch. Model-level insight shows that both absolute and relative temperature variations are important in solar panel inspections. The relative decrease in “blueness” provides a crucial early indication of faults, especially in low-contrast thermal images where distinguishing normal warm areas from actual hotspot is difficult. Feature-level insight highlights how subtle changes in color composition, particularly reductions in blue components, serve as early indicators of developing anomalies. Full article

To address the challenges brought by the high penetration of renewable energy in power systems, such as multi-scale dynamic interactions, high feature dimensionality, and limited model generalization, this paper proposes a transient stability assessment (TSA) method that combines temporal feature selection with deep learning-based modeling. First, a two-stage feature selection strategy is designed using the inter-class Mahalanobis distance and Spearman rank correlation. This helps extract highly discriminative and low-redundancy features from wide-area measurement system (WAMS) time-series data. Then, a parallel LSTM-Transformer architecture is constructed to capture both short-term local fluctuations and long-term global dependencies. A variational inference mechanism based on a Gaussian mixture model (GMM) is introduced to enable dynamic representations fusion and uncertainty modeling. A composite loss function combining improved focal loss and Kullback–Leibler (KL) divergence regularization is designed to enhance model robustness and training stability under complex disturbances. The proposed method is validated on a modified IEEE 39-bus system. Results show that it outperforms existing models in accuracy, robustness, interpretability, and other aspects. This provides an effective solution for TSA in power systems with high renewable energy integration. Full article

Fluids that exhibit self-rewetting properties, such as aqueous long-chain alcohol solutions, display a unique quadratic relationship between surface tension and temperature and are marked by a positive gradient. This characteristic leads to distinctive patterns of thermocapillary convection and associated interfacial dynamics, setting self-rewetting fluids apart from normal fluids (NFs). The potential to improve heat transfer using self-rewetting fluids (SRFs) is garnering interest for use in various technologies, including low-gravity conditions and microfluidic systems. Our research aims to shed light on the contrasting behaviors of SRFs in comparison to NFs regarding interfacial transport phenomena. This study focuses on the thermocapillary convection in SRF layers with a deformable interface enclosed inside a closed container modeled as a square cavity, which is subject to nonuniform heating, represented using a Gaussian profile for the heat flux variation on one of its sides, in the absence of gravity. To achieve this, we have enhanced a central-moment-based lattice Boltzmann method (LBM) utilizing three distribution functions for tracking interfaces, computing two-fluid motions with temperature-dependent surface tension and energy transport, respectively. Through numerical simulations, the impacts of several characteristic parameters, including the viscosity and thermal conductivity ratios, as well as the surface tension–temperature sensitivity parameters, on the distribution and magnitude of the thermocapillary-driven motion are examined. In contrast to that in NFs, the counter-rotating pair of vortices generated in the SRF layers, due to the surface tension gradient at the interface, is found to be directed toward the SRF layers’ hotter zones. Significant interfacial deformations are observed, especially when there are contrasts in the viscosities of the SRF layers. The thermocapillary convection is found to be enhanced if the bottom SRF layer has a higher thermal conductivity or viscosity than that of the top layer or when distributed, rather than localized, heating is applied. Furthermore, the higher the magnitude of the effect of the dimensionless quadratic surface tension sensitivity coefficient on the temperature, or of the effect of the imposed heat flux, the greater the peak interfacial velocity current generated due to the Marangoni stresses. In addition, an examination of the Nusselt number profiles reveals significant redistribution of the heat transfer rates in the SRF layers due to concomitant nonlinear thermocapillary effects. Full article

In this paper, for the needs of high-precision and high-continuity localization in complex environments, a modeling method based on time-varying noise and outlier noise is proposed, and variational Bayesian strong tracking filtering is used for 5G/INS fusion localization. A hierarchical progressive fault detection mechanism is proposed to detect IMU rationality faults and consistency faults in 5G observation information. The main contributions are reflected in the following two aspects: first, by innovatively introducing Pearson VII-type distribution for noise modeling, dynamically adjusting the tail thickness characteristics of the probability density function through its shape parameter, and effectively capturing the distribution law of extreme values in the observation data. Afterward, this article combined the variational Bayesian strong tracking filtering algorithm to construct a robust state estimation framework, significantly improving the localization accuracy and continuity in non-Gaussian noise environments. Second, a hierarchical progressive fault detection mechanism is designed. A wavelet fault detection method based on a hierarchical voting mechanism is adopted for IMU data to extract the abrupt features of the observed data and quickly identify faults. In addition, a dual-channel consistency detection model with dynamic fault-tolerant management was constructed. Sudden and gradual faults were quickly detected through a dual-channel pre-check, and then, the fault source was identified through AIME. Based on the fault source detection results, corresponding compensation mechanisms were adopted to achieve robust continuous localization. Full article

Hyperspectral image (HSI) quality is generally degraded by diverse noise contamination during acquisition, which adversely impacts subsequent processing performance. Current techniques predominantly rely on nuclear norms and low-rank matrix approximation theory to model the inherent property that HSIs lie in a low-dimensional subspace. Recent research has demonstrated that HSI gradient maps also exhibit low-rank priors. The correlated total variation (CTV), which is defined as the nuclear norm of gradient maps, can simultaneously model low-rank and local smoothness priors, and shows better performance than the standard nuclear norm. However, similar to nuclear norms, CTV may excessively penalize large singular values. To overcome these constraints, this study introduces a non-convex correlated total variation (NCTV), which shows the potential to eliminate mixed noise (including Gaussian, impulse, stripe, and dead-line noise) while preserving critical textures and spatial–spectral details. Numerical experiments on both simulated and real HSI datasets demonstrate that the proposed NCTV method achieves better performance in detail retention compared with the state-of-the-art techniques. Full article

Unbounded dynamic scene reconstruction is crucial for applications such as autonomous driving, robotics, and virtual reality. However, existing methods struggle to reconstruct dynamic scenes in unbounded outdoor environments due to challenges such as lighting variation, object motion, and sensor limitations, leading to inaccurate geometry and low rendering fidelity. In this paper, we proposed Gaussian-UDSR, a novel 3D Gaussian-based representation that efficiently reconstructs and renders high-quality, unbounded dynamic scenes in real time. Our approach fused LiDAR point clouds and Structure-from-Motion (SfM) point clouds obtained from an RGB camera, significantly improving depth estimation and geometric accuracy. To address dynamic appearance variations, we introduced a Gaussian color feature prediction network, which adaptively captures global and local feature information, enabling robust rendering under changing lighting conditions. Additionally, a pose-tracking mechanism ensured precise motion estimation for dynamic objects, enhancing realism and consistency. We evaluated Gaussian-UDSR on the Waymo and KITTI datasets, demonstrating state-of-the-art rendering quality with an 8.8% improvement in PSNR, a 75% reduction in LPIPS, and a fourfold speed improvement over existing methods. Our approach enables efficient, high-fidelity 3D reconstruction and fast real-time rendering of large-scale dynamic environments, while significantly reducing model storage overhead. Full article

Accurately simulating the thermal behavior of wheel–brake shoe friction on long downhill ramps is challenging due to the complexity of modeling appropriate heat source models. This study investigates heat generation during the frictional braking process of freight train wheels and brake shoes under long-slope conditions. Four heat source models—constant, modified Gaussian, sinusoidal, and parabolic distributions—were developed based on energy conservation principles and validated through experimental data. A thermomechanical coupled finite element model was established, incorporating a moving heat source to analyze the effects of different models on wheel tread temperature distribution and its evolution over time. The results show that all four models effectively simulate frictional heat generation, with computed temperatures, deviating by only 6.0–8.2% from experimental measurements, confirming their accuracy and reliability. Among the models, the modified Gaussian distribution heat source, with its significantly higher peak local heat flux (2.82 times that of the constant model) and rapid attenuation, offers the most precise simulation of the non-uniform temperature distribution in the contact region. This leads to a 40% increase in the temperature gradient variation rate and effectively reproduces the “hot spot” effect. The new non-uniform heat source model accurately captures local temperature dynamics and predicts frictional heat transfer and thermal damage trends. The modified Gaussian distribution model outperforms others in simulating local temperature peaks, offering support for optimizing braking system models and improving thermal damage prediction. Future research will refine this model by incorporating factors like material wear, environmental conditions, and dynamic contact characteristics. Full article

The score-based diffusion model has made significant progress in the field of computer vision, surpassing the performance of generative models, such as variational autoencoders, and has been extended to applications such as speech enhancement and recognition. This paper proposes a U-Net architecture using a score-based diffusion model and an efficient multi-scale attention mechanism (EMA) for the speech enhancement task. The model leverages the symmetric structure of U-Net to extract speech features and captures contextual information and local details across different scales using the EMA mechanism, improving speech quality in noisy environments. We evaluate the method on the VoiceBank-DEMAND (VB-DMD) dataset and the DARPA TIMIT Acoustic-Phonetic Continuous Speech Corpus–TUT Sound Events 2017 (TIMIT-TUT) dataset. The experimental results show that the proposed model performed well in terms of speech quality perception (PESQ), extended short-time objective intelligibility (ESTOI), and scale-invariant signal-to-distortion ratio (SI-SDR). Especially when processing out-of-dataset noisy speech, the proposed method achieved excellent speech enhancement results compared to other methods, demonstrating the model’s strong generalization capability. We also conducted an ablation study on the SDE solver and the EMA mechanism, and the results show that the reverse diffusion method outperformed the Euler–Maruyama method, and the EMA strategy could improve the model performance. The results demonstrate the effectiveness of these two techniques in our system. Nevertheless, since the model is specifically designed for Gaussian noise, its performance under non-Gaussian or complex noise conditions may be limited. Full article