Search Results (3,506)

Recent advances in ear biometrics have yielded increasingly accurate detection and recognition methods, driven by the ear’s uniqueness and permanence as a non-invasive biometric modality. Nonetheless, several limitations persist, including computationally demanding models, inconsistent evaluation metrics, and portable systems restricted by manual capture and limited datasets. To address these challenges, we developed a lightweight, video-based ear biometric system implemented on the Raspberry Pi 5. The system integrates You Only Look Once Version 12 (YOLOv12) for ear detection, EfficientNet-4 for feature extraction, and k-Nearest Neighbors (k-NNs) for recognition. Its robust hardware platform combines Raspberry Pi 5 with the Raspberry Pi AI Camera and AI HAT+. To train, fine-tune, and optimize YOLOv12 and EfficientNet-4, we used the Visual Geometry Group (VGG)Face-Ear dataset for training and the Unconstrained Ear Recognition Challenge 2019 dataset for validation, with k-NN employed for classification. The system is evaluated for classification accuracy and system-level performance. 13 participants, comprising 10 enrolled and three unenrolled subjects, participated in testing the system. The enrolled participants registered in the system were correctly identified, whereas unenrolled participants were excluded and rejected. The system achieved 92.31% accuracy, 95.45% precision, 96.97% recall, and an F1-score of 0.95, confirming the feasibility of deploying advanced ear biometric methods on embedded, resource-constrained devices. Full article

The high installation and operational costs of reference-grade air quality monitoring systems have accelerated the widespread adoption of low-cost sensors (LCS). However, their susceptibility to environmental influences, temporal drift, and measurement uncertainty necessitates robust calibration approaches to ensure reliable measurements. Although machine learning (ML)-based calibration methods have been widely investigated, most existing implementations rely on static analytical workflows and require programming expertise, which limits their accessibility for many domain specialists. To simplify and standardize the calibration process for low-cost air quality sensors, this study presents Air Quality Multi-Model Calibration (AQ-MultiCal), an interactive, no-code platform. The platform provides a unified environment for evaluating 14 regression models, performing automated hyperparameter optimization, and conducting comparative performance analysis through an intuitive graphical interface supported by interactive visualization tools. The platform is validated using CO₂ measurements collected from January and February 2025. Experimental results indicate that the optimized k-nearest neighbors (kNN) model achieved the best performance, with a coefficient of determination of R² = 0.990 with low prediction error. These results demonstrate that AQ-MultiCal enables accurate sensor calibration and systematic comparison of ML models while improving the accessibility of ML-based calibration through an open-source platform designed for domain experts without programming expertise. Full article

This study focuses on evaluating the effect of annealing time on the mechanical properties and structural changes in polyamide-based materials manufactured using fused filament fabrication (FFF) technology. Three materials were experimentally analysed: neat polyamide PA6, polyamide reinforced with 30% glass fibers (PA6 GF30), and the composite material Onyx. After fabrication, the test specimens were annealed at a temperature of 180 °C for 30, 60, and 100 min. Mechanical properties were evaluated by tensile testing in accordance with ISO 527, and the obtained data were further processed using machine learning methods (Linear SVM, Quadratic SVM, and K-NN) to classify individual levels of thermal exposure. The results showed that annealing significantly improved the tensile strength of Onyx from 50.78 ± 1.46 MPa (0 min) to 60.09 ± 1.30 MPa after 30 min, corresponding to an increase of approximately 18%, while further annealing (60 and 100 min) resulted in values between 59.23 and 62.12 MPa without statistically significant additional improvement. In contrast, PA6 GF30 exhibited a progressive decrease in tensile strength from 76.85 ± 0.87 MPa (0 min) to 51.91 ± 8.03 MPa after 100 min, representing an overall reduction of approximately 32%, indicating degradation of the polymer–fiber interface. For neat PA6, tensile strength decreased from 55.31 ± 3.83 MPa to 40.03 ± 9.36 MPa, but these differences were not statistically significant (p > 0.05). Machine learning classification confirmed predominantly linear material behavior, with Linear SVM achieving accuracies of 85% for Onyx and PA6, and 95% for PA6 GF30, outperforming Quadratic SVM and K-NN models. These findings provide valuable insights for optimizing post-processing conditions of FFF-manufactured polyamide materials and composites. Full article

The increasing prevalence of work-related stress requires advanced, non-intrusive physiological monitoring solutions. As conventional methods are often impractical for continuous, real-world applications, this study investigates the deployment of artificial intelligence models on embedded systems for real-time emotion recognition from physiological signals. The study identified critical constraints for embedded implementation, including model size and memory capacity. An evaluation of various machine learning algorithms revealed that, while models like K-Nearest Neighbors (KNN) achieve high accuracy (88.8%), their excessive memory footprints make them unsuitable for resource-constrained hardware. Consequently, Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and recurrent neural network (RNN) architectures were deployed on an STM32F411 microcontroller, for which model compression proved essential. An experimental study validated the approach, achieving high recognition rates for pronounced emotions such as hatred (91%) and anger (85%), though with a lower accuracy for more subtle states. These results confirm the potential of embedded AI systems for physiological monitoring, highlighting the critical importance of feature selection and model compression for practical implementation. Full article

The increasing scarcity of urban water resources, particularly in regions with intermittent supply and household water storage tanks, demands monitoring approaches capable of identifying end-use consumption patterns beyond aggregated volume measurements. Framed primarily as a feasibility study, this research presents an IoT-based framework for the automated classification of residential water consumption activities using water-level dynamics and supervised machine learning. A non-intrusive sensing architecture based on hydrostatic pressure measurements was deployed in a domestic water tank and integrated with a cloud-based data acquisition and processing platform. Five representative household states and activities were considered: tank refilling, stable state, toilet flushing, washing clothes, and taking a bath. A labeled dataset comprising 4396 consumption events was used to train and evaluate Decision Tree, Random Forest, Support Vector Machine (SVM), k-Nearest Neighbors, and Recurrent Neural Network (LSTM) models using features derived from water-level variations. All models achieved high performance, with accuracies above 0.92 and weighted F1-scores up to 0.93. The evaluated models showed highly comparable results, with the SVM (RBF) achieving a slightly higher accuracy (0.9307) in this evaluation setting, while ROC analysis showed AUC values between 0.97 and 1.00 across all classes, indicating strong discriminative capability. Additionally, specific activities such as washing clothes and tank refilling achieved precision and recall values above 0.95. These findings confirm that hydrostatic pressure-based sensing, combined with machine learning, enables reliable identification of domestic water-use events under intermittent supply conditions. The proposed approach provides actionable insights for demand management, leak detection, and user awareness, supporting more efficient and sustainable residential water consumption strategies. Full article

The objective of this research is to analyze the relationship between ESG factors and financial development, measured by Domestic Credit to the Private Sector by Banks (DCB). The empirical analysis employs a balanced panel of 82 countries for the years 2016 to 2022, obtained from the World Bank database. The proposed econometric model incorporates multiple ESG factors, including environmental (E), social (S), and governance (G). The list of econometric models under consideration includes fixed effects, random effects, WLS (weighted least squares), dynamic panel, and fixed effects with HAC estimation. Based on the conducted tests, the fixed effects estimation method has been chosen because the presence of serial correlation, heteroskedasticity, and cross-sectional dependence suggests that other methods will not provide an adequate model. As a result, fixed effects enable obtaining reliable estimates regarding the relationships between ESG factors and DCB. In addition, a KNN (K-Nearest Neighbors) regression was used to analyze potential nonlinear effects of the factors. The results show the strong positive relationship between ESG factors and financial development. More specifically, the presence of clean energy sources is associated with a positive DCB, and the depletion of natural resources is negatively associated with DCB. Moreover, social and governance factors are positively associated with financial development. Full article

Accurate prediction of the Remaining Useful Life (RUL) of lithium-ion batteries is essential for improving the reliability, safety, and maintenance planning of electric vehicles and energy storage systems. However, battery degradation is a complex and nonlinear process influenced by multiple operational conditions, making reliable RUL estimation a challenging task. Although numerous data-driven approaches have been proposed in the literature, many studies focus primarily on improving prediction accuracy using a single modeling technique, while limited attention has been given to systematic comparisons of different algorithms and the quantification of prediction uncertainty. This study proposes a comprehensive data-driven framework for lithium-ion battery RUL prediction by integrating both traditional machine learning and deep learning approaches. A Convolutional Neural Network (CNN) model is developed to capture nonlinear degradation patterns from battery cycling data. The dataset was divided using a battery-wise validation strategy to evaluate model generalization. In addition, conventional machine learning algorithms, including k-Nearest Neighbors (KNNs), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), are implemented to perform a comparative analysis of different predictive models. Key degradation-related features derived from voltage, current, temperature, and cycle information are extracted through a structured preprocessing pipeline. Furthermore, prediction uncertainty is quantified by constructing confidence intervals around the estimated RUL values. The predictive performance of the models is evaluated using prognostic metrics such as Root Mean Square Error (RMSE), Relative Prediction Error (RPE), and Prognostic Horizon (PH). The performance of the models is evaluated using multiple prognostic metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the coefficient of determination (R²), to ensure a comprehensive assessment of prediction accuracy. The experimental results demonstrate that the proposed framework provides accurate RUL predictions. Among the evaluated models, the CNN achieved the best performance with a Mean Absolute Error (MAE) of 7.75 and a Root Mean Square Error (RMSE) of 10.80, outperforming traditional machine learning models such as Random Forest and XGBoost. The KNN model also showed competitive performance with an RMSE of 12.07 and an R² value of 0.64, indicating that similarity-based learning can effectively capture battery degradation patterns. Full article

Milkfish (Chanos chanos), a widely consumed fish in the Philippines, is highly perishable, and conventional freshness assessments based on physical and olfactory inspection are often subjective and unreliable. To address this, we introduce an electronic nose system for the accurate classification of milkfish freshness based on spoilage-related gas emissions, namely methane, ammonia, hydrogen sulfide, and trimethylamine. The system integrates the MQ-series sensors and Taguchi gas sensor with Arduino Nano and Raspberry Pi 5 for data acquisition and signal processing. The k-nearest neighbor algorithm was used for classification, and its performance was evaluated using a confusion matrix. The data was gathered from 100 samples, consisting of 50 fresh and 50 spoiled fish. The evaluation demonstrated a peak classification accuracy of 92% for k-values between 1 and 15, confirming the system’s reliability. These findings indicate the system’s potential as a practical, low-cost, and efficient tool for enhancing consumer safety and quality assurance in the fish supply chain. Full article

The growing accumulation of microplastics in marine environments demands fast and accurate analytical methods for polymer identification. This study presents a new canonical spectral transformation (CST) strategy designed to extract the most relevant information of Raman spectra and enhance the performance of artificial intelligence (AI) models in the classification of microplastics. Using the Marine Plastic Database (MPDB) as the source of Raman spectra, five supervised models—k-Nearest Neighbor (KNN), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Multilayer Perceptron (MLP), and a one-dimensional Convolutional Neural Network (CNN-1D)—were trained and evaluated under both typical (conventional methodology) and CST workflows using 500 noisy samples per category. The CST consists of representing a Raman spectra in a vector where only the magnitude peaks of the most relevant frequency bands of the spectra are retained and the remaining values are null. This CST minimizes the inclusion of non-target data reaching the AI models. All models achieved higher accuracy with CST, where CNN-1D achieved the most significant performance, increasing accuracy to 0.90. In addition, CNN-1D identified Polystyrene (PS) and Poly(methyl methacrylate) (PMMA) with a score of 100% and 99%, respectively. The results demonstrate that CST effectively enhances spectral feature extraction and can be generalized to other spectroscopic techniques, providing a scalable framework for AI-assisted microplastic identification in seawater samples. Full article

This paper presents a physics-guided machine learning algorithm for classifying femur fracture presence and subtype using non-ionising radiofrequency (RF) spectral data. Multi-sensor S-parameter responses were generated from a femur phantom model across 1.0–3.0 GHz, producing 104 specimens representing intact bone and three fracture geometries. An exploratory, effect-size-driven band-selection algorithm identified a compact discriminative region between 1.74 and 1.90 GHz. Interpretable classifiers, including k-nearest neighbours (KNN), decision trees, linear discriminant analysis, and Naïve Bayes, were evaluated under strict specimen-level hold-out protocols to prevent data leakage. The KNN algorithm achieved 99.3% frame-level accuracy and 100% specimen-level accuracy for binary fracture detection while maintaining strong robustness in multiclass subtype classification, validated through sensor ablation and leave-one-subtype-out testing. The results demonstrate that compact, interpretable algorithms operating on band-limited RF spectra can achieve reliable, radiation-free fracture classification, supporting future development of continuous and edge-deployable monitoring systems. Full article

This study investigates the feasibility of acoustic signal analysis for the assessment of weld bead quality in the shielded metal arc welding (SMAW) process. The work focuses on comparing time-domain acoustic signals and time–frequency spectrogram representations for the classification of welds as accepted or rejected according to standard welding inspection criteria. Two key acoustic descriptors, the fundamental frequency (F₀) and the harmonics-to-noise ratio (HNR), were extracted and analyzed to evaluate statistical differences between the two weld quality classes. Statistical tests, including Anderson–Darling, Levene, ANOVA, and Kruskal–Wallis (α = 0.05), revealed significant differences between accepted and rejected welds. Accepted welds exhibited a bimodal HNR distribution associated with transient arc instability at the beginning and end of the bead, whereas rejected welds showed more uniform acoustic behavior throughout the process. Subsequently, the acoustic data were represented using both audio signals and spectrograms and used as inputs for ten supervised machine learning models, including Support Vector Classifier (SVC), Logistic Regression (LR), k-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Extra Trees (ET), Gradient Boosting (GB), and Naïve Bayes (NB). The results demonstrate that spectrogram-based representations significantly outperform time-domain signals, achieving accuracies of 0.95–0.96, ROC-AUC values above 0.95, and false positive and false negative rates below 6%. These findings indicate that, while scalar acoustic descriptors provide statistically significant insight into weld quality, time–frequency representations combined with machine learning enable a more robust and reliable framework for automated non-destructive evaluation, particularly in manual SMAW processes under realistic operating conditions. Full article

Background and Objectives: The prioritization of intensive care unit (ICU) admission following surgery for cancer is controversial. There is an urgent need to develop an appropriate clinical predictive model to aid in making ICU admission decisions after surgery. Materials and Methods: Four model strategies were used to build post−operative ICU admission predictive models: SVM, Catboost, ANN, and KNN. Internal verification was used for model evaluation at a ratio of 7:3. The area under the curve (AUC) value, calibration plots, and decision curve analysis were employed to assess the performance and clinical usefulness of the model. Results: The ICU group of patients with cancer who underwent surgery showed better prognosis for disease−free survival (DFS, p = 0.0008) and overall survival (OS, p < 0.0001). Cox multivariate analyses validated that lower baseline RBC, LDH, and CRP; higher baseline ALB, LCR, and lower post−operative LDH; higher post−operative HCT and ApoA−I; and higher fluctuating MCH independently predicted better DFS and OS (all p < 0.05). The AUC of the Catboost model was superior to that of the other models in the training cohort and internal validation cohort. Calibration plot and decision curve analysis indicated that the Catboost model possessed the best performance, with higher clinical utility, compared with other models. Conclusions: ICU admission after surgery was associated with superior survival in patients with cancer. The cost−effective Catboost model has promising clinical application but requires further clinical evaluation. Unravelling the cellular and molecular foundation of ICU admission might enable the development of more practical life−support strategies. Full article

In automotive Computer-Aided Engineering (CAE), the fidelity of high-quality shell element meshes is fundamentally governed by the accuracy of mid-surface geometry extraction. Conventional manual extraction for complex automotive plastic components is labor-intensive, error-prone, and often compromises mesh quality. To address these issues, this paper proposes an automated mid-surface mesh reconstruction method based on point cloud registration, establishing an integrated framework comprising “Multimodal Registration—Displacement Binding—Surface Correction.” Using a source part with an ideal mid-surface as a template, the method integrates Random Sample Consensus (RANSAC) and Iterative Closest Point (ICP) for rigid registration and Coherent Point Drift (CPD) for non-rigid registration to achieve high-precision alignment between the target and source outer-surface point clouds. Subsequently, a K-Nearest Neighbor (K-NN) search-based displacement binding mechanism smoothly transfers the outer-surface displacement field to the source mid-surface point cloud. Following position correction and surface smoothing, a complete and high-quality target mid-surface mesh is generated. Experimental results on typical plastic snap-fit components demonstrate that the normal projection error between the generated mid-surface and the manually refined “gold standard” mesh is less than 0.05 mm. The processing time per component is approximately 38 s, representing an efficiency improvement of over 73% compared to manual extraction using commercial CAE software. This method effectively mitigates common issues such as mid-surface distortion and feature loss, offering a high-precision, fully automated solution for automotive CAE pre-processing. Full article

The Welded Bead Bending Test (WBBT) assesses steel structures intended for construction in Germany in accordance with ZTV-ING Part 4 or DBS 918 002-02, as specified in Stahl-Eisen-Prüfblatt (SEP) 1390. Test outcomes are classified as passed (p) if the minimum bending angle $\alpha \ge {60}^{\circ }$ is achieved without fracture, not passed (n.p.) if fracture occurs beforehand, and invalid if no crack propagates into the base material. This study evaluates eight supervised machine learning models for classification regarding their suitability for predicting WBBT results: Decision Tree Classifier (DT), Random Forest Classifier (RF), Histogram-based Gradient Boosting Classifier (HGBC), k-Nearest-Neighbour (KNN), Bagging Classifiers based on DT (BCDT) and RF (BCRF), Generalized Learning Vector Quantizer (GLVQ), and Generalized Matrix Learning Vector Quantizer (GMLVQ). An industrial dataset of approximately 3600 samples was compiled in collaboration with Chemnitzer Werkstoff und Oberflächentechnik GmbH (CEWUS). Evaluation metrics included Balanced Accuracy, Recall, Specificity, computation time, and prediction stability. BCDT and BCRF achieved the highest Balanced Accuracy (70.6% and 70.3%, respectively), with BCRF excelling in Specificity (82.5%), thereby reliably detecting the n.p. class. GLVQ and GMLVQ demonstrated superior stability (maximum variability between training and testing dataset 0.14% and 3.17%, respectively), while BCRF and GMLVQ required the longest training times (BCRF: 10 s–20 s; GMLVQ: up to 80 s). KNN proved least suitable for WBBT outcome prediction. Full article

Background: Conventional analyses of physiological recovery often rely on discrete metrics that assume independence across time points, thereby ignoring intrinsic temporal continuity and masking substantial interindividual heterogeneity. This proof-of-concept study assesses the efficacy of Functional Data Analysis (FDA) as a promising framework for characterizing individual response dynamics following a functional threshold power (FTP) test. Methods: Physiological time-series data (including blood lactate, heart rate, blood pressure, and glucose levels) collected from 21 trained cyclists (10 professionals, 11 amateurs) were represented as functional objects using FDataGrid on the original sampling grid (0, 3, 5, 10, 20 min), without basis expansion or smoothing. We conducted unsupervised functional clustering (K-means; Fuzzy K-means) and supervised classification (Maximum Depth with Modified Band Depth, K-Nearest Neighbors, Nearest Centroid, functional QDA with parametric Gaussian covariance). Model performance was estimated via Repeated Stratified 5-Fold Cross-Validation with 10 repetitions (50 folds), reporting accuracy, balanced accuracy (mean ± SD), 95% CIs, permutation p-values, and sensitivity/specificity from aggregated confusion matrices. Results: Lactate (CL) and diastolic blood pressure (DBP) provided useful and statistically significant discrimination across several classifiers (e.g., KNN, Nearest Centroid, functional QDA), whereas heart rate showed modest discriminative value and glucose intermediate performance. Unsupervised analyses revealed distinct lactate recovery profiles and graded membership for hemodynamic/metabolic variables, supporting the value of FDA for resolving heterogeneity beyond group-average trends. Conclusions: FDA offers a feasible and informative approach for classifying recovery phenotypes while preserving temporal structure. Findings are promising but should be interpreted with caution due to the small sample size, sparse time points, and the need for external validation in larger, independent cohorts before translation into routine decision-making. Full article