Search Results (5,148)

This paper presents a novel hybrid and hierarchical architecture of a Convolutional Neural Network (CNN), based on MobileNetV2 and Support Vector Machines (SVM) for the classification of crop diseases (SVMobileNetV2). The system feeds from multispectral images captured by Unmanned Aerial Vehicles (UAVs) alongside data from IoT nodes. The primary objective is to improve classification performance in terms of both accuracy and precision. This is achieved by integrating contemporary Deep Learning techniques, specifically different CNN models, a prevalent type of artificial neural network composed of multiple interconnected layers, tailored for the analysis of agricultural imagery. The initial layers are responsible for identifying basic visual features such as edges and contours, while deeper layers progressively extract more abstract and complex patterns, enabling the recognition of intricate shapes. In this study, different datasets of tropical crop images, in this case banana crops, were constructed to evaluate the performance and accuracy of CNNs in detecting diseases in the crops, supported by transfer learning. For this, multispectral images are used to create false-color images to discriminate disease through spectra related to the blue, green and red colors in addition to red edge and near-infrared. Moreover, we used IoT nodes to include environmental data related to the temperature and humidity of the environment and the soil. Machine Learning models were evaluated and fine-tuned using standard evaluation metrics. For classification, we used fundamental metrics such as accuracy, precision, and the confusion matrix; in this study was obtained a performance of up to 86.5% using current deep learning models and up to 98.5% accuracy using the proposed hybrid and hierarchical architecture (SVMobileNetV2). This represents a new paradigm to significantly improve classification using the proposed hybrid CNN-SVM architecture and UAV-based multispectral images. Full article

Visual anomaly detection (VAD) plays a critical role in manufacturing and quality inspection, where the scarcity of anomalous samples poses challenges for developing reliable models. Existing approaches primarily rely on unsupervised training with synthetic anomalies, which often favor specific defect types and struggle to generalize across diverse categories. To address these limitations, we propose MADE-Net (Multi-model Adaptive anomaly Detection Ensemble Network), an industrial anomaly detection framework that integrates three complementary submodels: a reconstruction-based submodel (SRAD), a feature embedding-based submodel (SFAD), and a patch discrimination submodel (LPD). A dynamic integration and selection module (ISM) adaptively determines the most suitable submodel output according to input characteristics. We further introduce ManuDefect-21, a large-scale benchmark dataset comprising 11 categories of electronic components with both normal and anomalous samples in the training and test sets. The dataset reflects realistic positive-to-negative ratios and diverse defect types encountered in real manufacturing environments, addressing several limitations of previous datasets such as MVTec-AD and VisA. Experiments conducted on ManuDefect-21 demonstrate that MADE-Net achieves consistent improvements in both detection and localization metrics (e.g., average AUROC of 98.5%, Pixel-AP of 68.7%) compared with existing methods. While MADE-Net requires pixel-level annotations for fine-tuning and introduces additional computational overhead, it provides enhanced adaptability to complex industrial conditions. The proposed framework and dataset jointly contribute to advancing practical and reproducible research in industrial anomaly detection. Full article

Background/Objectives: Peritonsillar abscess (PTA) and cellulitis (PTC) often present with similar clinical features, making differentiation challenging despite imaging. This study evaluates the diagnostic performance of serum HMGB1 and kallistatin levels as potential independent biomarkers to distinguish PTA from PTC. Methods: In this single-center prospective cohort study, 97 patients aged 18 to 65 years who met the inclusion criteria and presented with peritonsillar infection (39 PTA; 58 PTC) between February and July 2025 were enrolled. Serum levels of HMGB1, kallistatin, and routine inflammatory markers were measured and compared. Univariate and multivariate logistic regression analyses identified independent predictors for distinguishing PTA from PTC. Receiver operating characteristic (ROC) curve analysis assessed the diagnostic accuracy of biomarkers. Decision curve analysis (DCA) was performed to evaluate the clinical net benefit of individual biomarkers and their combinations across a range of threshold probabilities. Results: Compared to controls, patients with peritonsillar infection had significantly higher WBC, neutrophil, CRP, procalcitonin, and HMGB1 levels and significantly lower kallistatin levels (all p < 0.05). Within the infection group, PTA patients showed significantly higher CRP (p = 0.036) and HMGB1 (p = 0.003) levels and lower kallistatin (p < 0.001) levels compared to PTC patients. In univariate analysis, CRP, HMGB1, and kallistatin were significantly associated with PTA; however, in multivariate analysis, only elevated HMGB1 (OR: 1.21; 95% CI: 1.09–1.35; p < 0.001) and reduced kallistatin (OR: 0.395; 95% CI: 0.24–0.648; p < 0.001) remained independent predictors. ROC analysis showed that both HMGB1 and kallistatin demonstrated good discriminative ability in distinguishing PTA from PTC. DCA revealed that the three-biomarker combination (kallistatin + HMGB1 + CRP) achieved the highest mean net benefit (0.183) across all threshold probabilities, outperforming individual biomarkers (kallistatin: 0.131, HMGB1: 0.111, CRP: 0.099) and the two-biomarker model (0.176). The combined model maintained superior net benefit across threshold probabilities of 25–75%, indicating optimal clinical utility within this decision range. Conclusions: Serum HMGB1 and kallistatin may be effective adjunctive biomarkers for differentiating PTA from PTC. Full article

The desert riparian forest oasis, dominated by Populus euphratica and Tamarix chinensis, is an important barrier to protect the economic production and habitat of the Tarim River Basin. However, there is still a lack of high-precision spatial distribution data of desert ri-parian forest species below 10 m. The recently launched PlanetScope CubeSat constella-tion, which provides daily earth observation imagery with a resolution of 3 m, offers a highly favorable dataset for mapping the high-resolution distribution of P. euphratica and T. chinensis and an unprecedented opportunity to explore the optimal phenology window to distinguish between them. In this study, time-series PlanetScope images were first used to extract phenological metrics of P. euphratica, dividing the annual life cycle into four phenology windows: duration of leaf expansion (DLE), duration of leaf maturity (DLM), duration of leaf fall (DLF), and duration of the dormancy period (DDP). The random forest model was used to obtain the classification accuracy of 16 phenological window combinations. Results indicate that after gap filling of vegetation index time series, the identification accuracy for P. euphratica and T. chinensis exceeded 0.90. Among individual phenology windows, the DLE window exhibited the highest classification accuracy (average F1-score 0.87). Among the two phenology window combinations, the DLE-DLF and DLE-DLM windows have the highest classification accuracy (average F1-score 0.90). Among the three phenology window combinations, DLE-DLM-DLF displayed the highest classification accuracy (average F1-score 0.91). Nevertheless, the inclusion of features within the DDP window led to a decrease in accuracy by 1–2% points, which was unfavorable for discriminating tree species. Additionally, features observed during the phenology asynchrony period were found to be more valuable for distinguishing between tree species. Our findings highlight the potential of PlanetScope constellation imagery in tree species classification, offering guidance for selecting optimal image acquisition timing and identifying the most valuable images within time series data for future large-scale tree mapping. Full article

As the core refrigeration equipment in cooling systems, the operational state of chiller units is crucial for ship support, equipment cooling, and mission stability. However, because of their sensitivity and the complexity of operating environments, obtaining large volumes of complete, fault-labeled data is difficult in practical engineering appli-cations. This limitation makes it challenging for traditional data-driven approaches to deliver accurate fault diagnoses. Furthermore, data collected from different devices or under varying operating conditions often differ significantly in both feature dimensions and distributions, i.e., data heterogeneity, which further complicates model transfer. To address these challenges, this study proposes a deep transfer learning–based fault di-agnosis method designed to leverage abundant knowledge from the source domain while adaptively learning features of the target domain. Given the persistent difficulties in collecting sufficient high-quality labeled fault data, traditional data-driven models continue to face restricted diagnostic performance on target equipment. At the same time, data heterogeneity across devices or operating conditions intensifies the challenge of cross-domain knowledge transfer. To overcome these issues, this study develops a heterogeneous transfer learning method that integrates a dual-channel autoencoder, domain adversarial training, and pseudo-label self-training. This combination enables precise small-sample knowledge transfer from the source to the target domain. Specifi-cally, the dual-channel autoencoder is first applied to align heterogeneous feature di-mensions. Then, a Gradient Reversal Layer (GRL) and a domain discriminator are in-troduced to extract domain-invariant features. In parallel, high-confidence pseu-do-labeled samples from the target domain are incorporated into joint training to im-prove generalization and robustness. Experimental results confirm that the method achieves high fault diagnosis accuracy in typical industrial application scenarios, ena-bling effective identification of common faults in various types of chiller units under conventional operating conditions, the proposed method achieves higher accuracy and F1-scores in multi-class fault diagnosis tasks compared with both traditional approaches and existing transfer learning methods. These findings provide a novel perspective for advancing the intelligent operation and maintenance of chiller units. Full article

Background/Objectives: Radiomics enables extraction of quantitative imaging features to support non-invasive classification of prostate cancer (PCa). Accurate detection of clinically significant PCa (csPCa; Gleason score ≥ 3 + 4) is crucial for guiding treatment decisions. However, many studies explore limited feature selection, classifier, and harmonization combinations, and lack external validation. We aimed to systematically benchmark modeling pipelines and evaluate whether combining radiomics with the lesion-to-normal ADC ratio improves classification robustness and generalizability in multicenter datasets. Methods: Radiomic features were extracted from ADC maps using IBSI-compliant pipelines. Over 100 model configurations were tested, combining eight feature selection methods, fifteen classifiers, and two harmonization strategies across two scenarios: (1) repeated cross-validation on a multicenter dataset and (2) nested cross-validation with external testing on the PROSTATEx dataset. The ADC ratio was defined as the mean lesion ADC divided by contralateral normal tissue ADC, by placing two identical ROIs in each side, enabling patient-specific normalization. Results: In Scenario 1, the best model combined radiomics, ADC ratio, LASSO, and Naïve Bayes (AUC-PR = 0.844 ± 0.040). In Scenario 2, the top-performing configuration used Recursive Feature Elimination (RFE) and Boosted GLM (a generalized linear model trained with boosting), generalizing well to the external set (AUC-PR = 0.722; F1 = 0.741). ComBat harmonization improved calibration but not external discrimination. Frequently selected features were texture-based (GLCM, GLSZM) from wavelet- and LoG-filtered ADC maps. Conclusions: Integrating radiomics with the ADC ratio improves csPCa classification and enhances generalizability, supporting its potential role as a robust, clinically interpretable imaging biomarker in multicenter MRI studies. Full article

In the interior decoration panel industry, automated production lines have become the standard configuration for large-scale enterprises. However, during the panel processing procedures such as sanding and painting, the loss of traditional identification markers like QR codes or barcodes is inevitable. This creates a critical technical bottleneck in the assembly stage of customized or multi-model parallel production lines, where identifying individual panels significantly limits production efficiency. To address this issue, this paper proposes a high-precision measurement method based on close-range photogrammetry for capturing panel dimensions and hole position features, enabling accurate extraction of identification markers. Building on this foundation, an identity discrimination method that integrates weighted dimension and hole position IDs has been developed, making it feasible to efficiently and automatically identify panels without physical identification markers. Experimental results demonstrate that the proposed method exhibits significant advantages in both recognition accuracy and production adaptability, providing an effective solution for intelligent manufacturing in the home decoration panel industry. Full article

We address issues in existing house layout generation methods, including chaotic room layouts, limited iterative refinement, and restricted style diversity. We propose Graph-RWGAN, a generative adversarial network based on a multi-relational graph attention mechanism, to automatically generate reasonable and globally consistent house layouts under weak constraints. In our framework, rooms are represented as graph nodes with semantic attributes. Their spatial relationships are modeled as edges. Optional room-level objects can be added by augmenting node attributes. This allows for object-aware layout generation when needed. The multi-relational graph attention mechanism captures complex inter-room relationships. Iterative generation enables stepwise layout optimization. Fusion of node features with building boundaries ensures spatial accuracy and structural coherence. A conditional graph discriminator with Wasserstein loss constrains global consistency. Experiments on the RPLAN dataset show strong performance. FID is 92.73, SSIM is 0.828, and layout accuracy is 85.96%. Room topology accuracy reaches 95%, layout quality 90%, and structural coherence 95%, outperforming House-GAN, LayoutGAN, and MR-GAT. Ablation studies confirm the effectiveness of each key component. Graph-RWGAN shows strong adaptability, flexible generation under weak constraints, and multi-style layouts. It provides an efficient and controllable scheme for intelligent building design and automated planning. Full article

The traditional Mahalanobis–Taguchi System (MTS) employs two main strategies for multi-class classification: the partial binary tree MTS (PBT-MTS) and the multi-tree MTS (MT-MTS). The PBT-MTS relies on a fixed binary tree structure, resulting in limited model flexibility, while the MT-MTS suffers from its dependence on pre-defined category partitioning. Both methods exhibit constraints in adaptability and classification efficiency within complex data environments. To overcome these limitations, this paper proposes an innovative Adaptive Tree-structured Mahalanobis–Taguchi System (ATMTS). Its core breakthrough lies in the ability to autonomously construct an optimal multi-layer classification tree structure. Unlike conventional PBT-MTS, which establishes a Mahalanobis Space (MS) containing only a single category per node, ATMTS dynamically generates the MS that incorporates multiple categories, substantially enhancing discriminative power and structural adaptability. Furthermore, compared to MT-MTS, which depends on prior label information, ATMTS operates without predefined categorical assumptions, uncovering discriminative relationships solely through data-driven learning. This enables broader applicability and stronger generalization capability. By introducing a unified multi-objective joint optimization model, our method simultaneously optimizes structure construction, feature selection, and threshold determination, effectively overcoming the drawbacks of conventional phased optimization approaches. Experimental results demonstrate that ATMTS outperforms PBT-MTS, MT-MTS, and other mainstream classification methods across multiple benchmark datasets, achieving significant improvements in the accuracy and robustness of multi-class classification tasks. Full article

Few-shot image classification seeks to recognize novel categories from only a handful of labeled examples, but conventional metric-based methods that rely mainly on global image features often produce unstable prototypes under extreme data scarcity, while local-descriptor approaches can lose context and suffer from inter-class local-pattern overlap. To address these limitations, we propose a Global–Local Feature Fusion network that combines a frozen, pretrained global feature branch with a self-attention based multi-local feature fusion branch. Multiple random crops are encoded by a shared backbone (ResNet-12), projected to Query/Key/Value embeddings, and fused via scaled dot-product self-attention to suppress background noise and highlight discriminative local cues. The fused local representation is concatenated with the global feature to form robust class prototypes used in a prototypical-network style classifier. On four benchmarks, our method achieves strong improvements: Mini-ImageNet 70.31% ± 0.20 (1-shot)/85.91% ± 0.13 (5-shot), Tiered-ImageNet 73.37% ± 0.22/87.62% ± 0.14, FC-100 47.01% ± 0.20/64.13% ± 0.19, and CUB-200-2011 82.80% ± 0.18/93.19% ± 0.09, demonstrating consistent gains over competitive baselines. Ablation studies show that (1) naive local averaging improves over global-only baselines, (2) self-attention fusion yields a large additional gain (e.g., +4.50% in 1-shot on Mini-ImageNet), and (3) concatenating global and fused local features gives the best overall performance. These results indicate that explicitly modeling inter-patch relations and fusing multi-granularity cues produces markedly more discriminative prototypes in few-shot regimes. Full article

Photovoltaic (PV) fault detection faces significant challenges in distinguishing subtle defects from complex backgrounds while maintaining reliability across diverse environmental conditions. Traditional approaches struggle with scalability and accuracy limitations, particularly when detecting electrical damage, physical defects, and environmental soiling in thermal imagery. This paper presents SolarFaultAttentionNet, a novel dual-attention deep learning framework that integrates channel-wise and spatial attention mechanisms within a multi-path CNN architecture for enhanced PV fault classification. The approach combines comprehensive data augmentation strategies with targeted attention modules to improve feature discrimination across six fault categories: Electrical-Damage, Physical-Damage, Snow-Covered, Dusty, Bird-Drop, and Clean. Experimental validation on a dataset of 885 images demonstrates that SolarFaultAttentionNet achieves 99.14% classification accuracy, outperforming state-of-the-art models by 5.14%. The framework exhibits perfect detection for dust accumulation (100% across all metrics) and robust electrical damage detection (99.12% F1 score) while maintaining an optimal sensitivity (98.24%) and specificity (99.91%) balance. The computational efficiency (0.0160 s inference time) and systematic performance improvements establish SolarFaultAttentionNet as a practical solution for automated PV monitoring systems, enabling reliable fault detection critical for maximizing energy production and minimizing maintenance costs in large-scale solar installations. Full article

Migraine is a common neurological illness that has a major influence on the quality of life; yet, precise categorization and prediction remain difficult because of its complicated symptoms and multiple triggers. This work investigates the use of advanced machine learning (ML) algorithms to improve migraine diagnosis and prediction, drawing on a large dataset that includes clinical, lifestyle, and environmental aspects. Various machine learning models, such as ensemble methods, deep learning, and hybrid approaches, are tested to see how well they discriminate migraine from other headache conditions and predict migraine episodes. Feature selection approaches are used to identify the most important predictors, which improve model interpretability and performance. Experimental results show that the proposed machine learning framework outperforms established diagnostic methods in terms of classification accuracy, sensitivity, and specificity. The study demonstrates how ML-driven solutions may be used to manage migraines in a tailored way, helping medical practitioners with early diagnosis and intervention techniques. My suggested framework, NeuroVote(ensemble model), offers a remarkable 99.99% classification accuracy for migraines. Future studies will concentrate on optimizing models for clinical deployment and incorporating real-time data from wearable technology. Full article

An Adaptive Neuro-Fuzzy Inference System (ANFIS) framework for paediatric wrist injury classification (fracture versus sprain) was developed utilising infrared thermography (IRT). ANFIS combines artificial neural network (ANN) learning with interpretable fuzzy rules, mitigating the “black-box” limitation of conventional ANNs through explicit membership functions and Takagi–Sugeno rule consequents. Forty children (19 fractures, 21 sprains, confirmed by X-ray radiograph) provided thermal image sequences from which three statistically discriminative temperature distribution features namely standard deviation, inter-quartile range (IQR) and kurtosis were selected. A five-layer Sugeno ANFIS with Gaussian membership functions were trained using a hybrid least-squares/gradient descent optimisation and evaluated under three premise-parameter initialisation strategies: random seeding, K-means clustering, and fuzzy C-means (FCM) data partitioning. Five-fold cross-validation guided the selection of membership functions standard deviation (σ) and rule count, yielding an optimal nine-rule model. Comparative experiments show K-means initialisation achieved the best balance between convergence speed and generalisation versus slower but highly precise random initialisation and rapidly convergent yet unstable FCM. The proposed K-means–driven ANFIS offered data-efficient decision support, highlighting the potential of thermal feature fusion with neuro-fuzzy modelling to reduce unnecessary radiographs in emergency bone fracture triage. Full article

This study pioneers a fully remote sensing-based framework for mapping heatwave susceptibility, integrating the TabTransformer deep learning model with Particle Swarm Optimization (PSO) for robust hyperparameter tuning. The central question addressed is whether a fully remote sensing-driven, PSO-optimized TabTransformer can achieve accurate, scalable, and spatially detailed heatwave susceptibility mapping in data-scarce regions such as Central Asia. Utilizing ERA5-derived heatwave evidence and thirteen environmental and socio-economic predictors, the workflow produces high-resolution susceptibility maps spanning five Central Asian countries. Comparative analysis evidences that the PSO-optimized TabTransformer model outperforms the baseline across multiple metrics. On the test set, the optimized model achieved an RMSE of 0.123, MAE of 0.034, and R² of 0.938, outperforming the standalone TabTransformer (RMSE = 0.132, MAE = 0.038, R² = 0.93). Discriminative capacity also improved, with AUROC increasing from 0.933 to 0.940. The PSO-tuned model delivered faster convergence, lower final loss, and more stable accuracy during training and validation. Spatial outputs reveal heightened susceptibility in southern and southwestern sectors—Turkmenistan, Uzbekistan, southern Kazakhstan, and adjacent lowlands—with statistically significant improvements in spatial precision and class delineation confirmed by Chi-squared, Friedman, and Wilcoxon tests, all with congruent p-values of <0.0001. Feature importance analysis consistently identifies maximum temperature, frequency of hot days, and rainfall as dominant predictors. These advancements validate the potential of data-driven, deep learning approaches for reliable, scalable environmental hazard assessment, crucial for climate adaptation planning in vulnerable regions. Full article

Psychogenic non-epileptic seizures (PNES) are often misdiagnosed as epileptic seizures (ES), leading to inappropriate treatment and delayed psychological care. To address this challenge, we analysed electroencephalogram (EEG) data from 74 patients (46 PNES, 28 ES) using one-minute preictal and interictal recordings per subject. Nine entropy measures (Sample, Fuzzy, Permutation, Dispersion, Conditional, Phase, Spectral, Rényi, and Wavelet entropy) were evaluated individually to classify PNES from ES using k-nearest neighbours, Naïve Bayes, linear discriminant analysis, logistic regression, support vector machine, random forest, multilayer perceptron, and XGBoost within a leave-one-subject-out cross-validation framework. In addition, a dynamic state, defined as the entropy difference between interictal and preictal periods, was examined. Sample, Fuzzy, Conditional, and Dispersion entropy were higher in PNES than in ES during interictal recordings (not significant), but significantly lower in the preictal (p < 0.05) and dynamic states (p < 0.01). Spatial mapping and permutation-based importance analyses highlighted O1, O2, T5, F7, and Pz as key discriminative channels. Classification performance peaked in the dynamic state, with Fuzzy entropy and support vector machine achieving the best results (balanced accuracy = 72.4%, F1 score = 77.8%, sensitivity = 74.5%, specificity = 70.4%). These results demonstrate the potential of entropy features for differentiating PNES from ES. Full article