Search Results (12,628)

Surface defect detection in aluminum-based composite core conductors (ACCC) via X-ray imaging has long been constrained by challenges such as small sample sizes, class imbalance, model redundancy, and inadequate adaptation to single-channel industrial images. To address this, this paper proposes SE-ResNet18, a lightweight classification model synergistically designed for industrial single-channel X-ray images. The model features a co-adapted architecture where a single-channel input layer (preserving native image information and eliminating RGB conversion overhead) is coupled with a channel attention mechanism (to amplify subtle defect features), all within a globally optimized lightweight framework. With targeted data augmentation and robust training strategies, the model achieves superior performance on the ACCC defect dataset: classification accuracy reaches 98.39%, while excelling in lightweight design (12.0 million parameters) and real-time capability (0.44 ms/image inference speed). The experiments demonstrate that the proposed model exhibits high classification accuracy in testing while offering superior lightweight characteristics and inference efficiency. This provides a feasible solution for achieving high-precision detection and real-time processing in industrial scenarios, showcasing potential for ACCC online detection applications. Full article

Dysarthria in patients post-stroke is often accompanied by central facial paralysis, which impairs facial motor control and emotional expression. Current assessments rely on acoustic modalities, overlooking facial pathological cues and their correlation with emotional expression, which hinders comprehensive disease assessment. To address this issue, we propose a multimodal severity classification framework that integrates facial and acoustic features. Firstly, a multi-level annotation algorithm based on a pre-trained model and motion amplitude was designed to overcome the problem of data scarcity. Secondly, facial topology was modeled using Delaunay triangulation, with spatial relationships captured via graph convolutional networks (GCNs), while abnormal muscle coordination is quantified using facial action units (AUs). Finally, we proposed a multimodal feature set fusion technology framework to achieve the compensation of facial visual features for acoustic modalities and the analysis of disease classification. Our experimental results using the THE-POSSD dataset demonstrate an accuracy of 92.0% and an F1 score of 91.6%, significantly outperforming single-modality baselines. This study reveals the changes in facial movements and sensitive areas of patients under different emotional states, verifies the compensatory ability of visual patterns for auditory patterns, and demonstrates the potential of this multimodal framework for objective assessment and future clinical applications in speech disorders. Full article

Decoding reaching movements from non-invasive brain signals is a key challenge for the development of naturalistic brain–computer interfaces (BCIs). While this decoding problem has been addressed via traditional machine learning, the exploitation of deep learning is still limited. Here, we evaluate a convolutional neural network (CNN) for decoding movement direction during a delayed center-out reaching task from the EEG. Signals were collected from twenty healthy participants and analyzed using EEGNet to discriminate reaching endpoints in three scenarios: fine-direction (five endpoints), coarse-direction (three endpoints), and proximity (two endpoints) classifications. To interpret the decoding process, the CNN was coupled with explanation techniques, including DeepLIFT and occlusion tests, enabling a data-driven analysis of spatio-temporal EEG features. The proposed approach achieved accuracies well above chance, with accuracies of 0.45 (five endpoints), 0.64 (three endpoints) and 0.70 (two endpoints) on average across subjects. Explainability analyses revealed that directional information is predominantly encoded during movement preparation, particularly in parietal and parietal–occipital regions, consistent with known visuomotor planning mechanisms and with EEG analysis based on event-related spectral perturbations. These results demonstrate the feasibility and interpretability of CNN-based EEG decoding for reaching movements, providing insights relevant for both neuroscience and the prospective development of non-invasive BCIs. Full article

Waste printed circuit boards (WPCBs) contain valuable metals such as gold, palladium, and silver, which are typically recovered through non-ferrous metal smelting. Currently, WPCBs are manually classified by workers, who visually compare board colors and component layouts with previously processed boards. This approach is time-consuming and prone to human error. To address these limitations, we propose an image-based algorithm for automated WPCB similarity assessment. The method extracts visual features from board images and computes similarity scores, incorporating classification strategies based on board-specific characteristics. Key features identified as effective for similarity evaluation include the hue value, coefficient of variation in terminal regions, number of line elements in terminal regions, structural complexity, and number of integrated circuits. Weighted feature contributions further improve accuracy. Our experimental results demonstrate that the proposed approach achieves 88.0% accuracy for the targeted PCB types, outperforming a comparative self-supervised contrastive learning method. This image-driven solution can significantly streamline WPCB recycling by reducing reliance on manual inspection and improving operational efficiency. Full article

Background/Objectives: The earlier, more accurate, and more consistent prediction of the brain tumor recognition process requires automated systems to minimize diagnostic delays and human error. The automated system provides a platform for handling large medical images, speeding up clinical decision-making. However, the existing system is facing difficulties due to the high variability in tumor location, size, and shape, which leads to segmentation complexity. In addition, glioma-related tumors infiltrate the brain tissues, making it challenging to identify the exact tumor region. Method: The above-identified research difficulties are overcome by applying the Swin-UNet with cuttlefish-optimized attention-based Graph Neural Networks (SCAG-Net), thereby improving overall brain tumor recognition accuracy. This integrated approach is utilized to address infiltrative gliomas, tumor variability, and feature redundancy issues by improving diagnostic efficiency. Initially, the collected MRI images are processed using the Swin-UNet approach to identify the region, minimizing prediction error robustly. The region’s features are explored utilizing the cuttlefish algorithm, which minimizes redundant features and speeds up classification by improving accuracy. The selected features are further processed using the attention graph network, which handles structural and heterogeneous information across multiple layers, improving classification accuracy compared to existing methods. Results: The efficiency of the system, implemented with the help of public datasets such as BRATS 2018, BRATS 2019, BRATS 2020, and Figshare is ensured by the proposed SCAG-Net approach, which achieves maximum recognition accuracy. The proposed system achieved a Dice coefficient of 0.989, an Intersection over Union of 0.969, and a classification accuracy of 0.992. This performance surpassed the most recent benchmark models by margins of 1.0% to 1.8% and with statistically significant differences (p < 0.05). These findings present a statistically validated, computationally efficient, clinically deployable framework. Conclusions: The effective analysis of MRI complex structures is used in medical applications and clinical analysis. The proposed SCAG-Net framework significantly improves brain tumor recognition by addressing tumor heterogeneity and infiltrative gliomas using MRI images. The proposed approach provides a robust, efficient, and clinically deployable solution for brain tumor recognition from MRI images, supporting accurate and rapid diagnosis while maintaining expert-level performance. Full article

To address the strong nonstationarity and complex multi-source interference in vibration signals of bogie axle-box bearings, a fault diagnosis method combining Improved Empirical Wavelet Transform (IEWT) and a Convolutional Neural Network (CNN)–Transformer model is proposed. First, the vibration signals are decomposed using the IEWT method, where dynamic frequency-band division adaptively determines the decomposition bands. This yields multiple intrinsic mode functions, and key modes containing fault features are selected based on information entropy. Next, the selected key modes are fused and transformed into polar coordinate projection maps, further enhancing the distinctiveness of fault data features. Finally, CNN is employed to extract local features from the vibration signals, while the Transformer captures long-range dependencies through the self-attention mechanism, significantly improving feature modeling for complex signals. To validate the fault diagnosis performance of the IEWT and CNN–Transformer model, vibration signals from bogie axle-box bearings in urban railways are analyzed. Analysis of the experimental data suggests that the adaptive decomposition of bearing signals using IEWT effectively overcomes the fixed band boundary limitations of traditional EWT, enhancing the precision of signal feature extraction. The integration of polar coordinate projection maps more accurately illustrates frequency variations and amplitude differences in the signals, fully capturing their nonstationary characteristics. Among the five fault categories of bogie axle-box bearings, the proposed method achieves an accuracy of 99.46%, a recall rate of 99.52%, and an F1-score of 0.995, significantly outperforming five classic comparison methods. This demonstrates that the combined strengths of CNN and Transformer yield higher classification accuracy and better robustness in handling complex fault patterns, effectively solving the fault diagnosis challenges for bogie axle-box bearings. Full article

Misdiagnosing plant diseases often leads to a range of negative consequences, including the overuse of pesticides and unnecessary food waste. Traditionally, identifying diseases on plant leaves has relied on manual visual inspection, making it a complex and time-consuming task. Since the advent of convolutional neural networks, however, recognition performance for leaf diseases has improved significantly. Most contemporary studies that apply AI techniques to plant-leaf disease classification focus primarily on boosting accuracy, frequently overlooking the limitations posed by resource-constrained real-world environments. To address these challenges, this thesis employs knowledge distillation to enable small models to approximate the recognition capabilities of larger ones. We enhance a ShuffleNetV2-based model by integrating multi-scale feature extraction and a coordinate-attention mechanism, and we further improve the lightweight student model through knowledge distillation to boost its recognition performance. Experimental results show that the proposed model achieves 93.15% accuracy on the Plant Pathology 2021- FGVC8 dataset, utilizing only 0.36 M parameters and 0.0931 GFLOPs. Compared to the ResNet50 baseline, our architecture slashes parameters by nearly 98% while limiting the accuracy gap to a mere 1.6%. These results confirm the model’s ability to maintain robust performance with minimal computational overhead, providing a practical solution for precision agriculture on resource-limited edge devices. Full article

Background and Objectives: Classification of thyroid nodules (TN) in ultrasound remains challenging due to limited labelled data and the limited capacity of conventional feature representations to capture complex, multi-directional textures. This work aims to improve data-efficient TN classification by integrating a geometry-adaptive Bandelet Transform (BT) with transfer learning (TL) to enhance feature representation and generalisation. Methods: The proposed pipeline first applies BT to strengthen directional and structural encoding in ultrasound images via quadtree-driven geometric adaptation. It then mitigates class imbalance using SMOTE and increases data diversity through targeted data augmentation. The resulting representations are classified using multiple ImageNet-pretrained architectures, where VGG19 yields the most consistent performance. Results: Experiments on the publicly available DDTI dataset show that BT-based preprocessing consistently improves performance over classical wavelet representations across multiple quadtree thresholds, with the best results obtained at $T=30$. Under this setting, the proposed BT+TL (VGG19) model achieves 98.91% accuracy, 98.11% sensitivity, 97.31% specificity, and a 98.89% F1-score, outperforming comparable approaches reported in the literature. Conclusions: Coupling geometry-adaptive transforms with modern TL backbones provides a robust and data-efficient strategy for ultrasound TN classification, particularly under limited annotation and challenging texture variability. The complete project is publicly available. Full article

Autism Spectrum Disorder (ASD) is a neurological condition that affects communication and social interaction skills, with individuals experiencing a range of challenges that often require specialized care. Automated systems for recognizing ASD face significant challenges due to the complexity of identifying distinguishing features from facial images. This study proposes an incremental advancement in ASD recognition by introducing a dual-stream model that combines handcrafted facial-landmark features with deep learning-based pixel-level features. The model processes images through two distinct streams to capture complementary aspects of facial information. In the first stream, facial landmarks are extracted using MediaPipe (v0.10.21),with a focus on 137 symmetric landmarks. The face’s position is adjusted using in-plane rotation based on eye-corner angles, and geometric features along with 52 blendshape features are processed through Dense layers. In the second stream, RGB image features are extracted using pre-trained CNNs (e.g., ResNet50V2, DenseNet121, InceptionV3) enhanced with Squeeze-and-Excitation (SE) blocks, followed by feature refinement through Global Average Pooling (GAP) and DenseNet layers. The outputs from both streams are fused using weighted concatenation through a softmax gate, followed by further feature refinement for classification. This hybrid approach significantly improves the ability to distinguish between ASD and non-ASD faces, demonstrating the benefits of combining geometric and pixel-based features. The model achieved an accuracy of 96.43% on the Kaggle dataset and 97.83% on the YTUIA dataset. Statistical hypothesis testing further confirms that the proposed approach provides a statistically meaningful advantage over strong baselines, particularly in terms of classification correctness and robustness across datasets. While these results are promising, they show incremental improvements over existing methods, and future work will focus on optimizing performance to exceed current benchmarks. Full article

Background/Objectives: Parkinson’s disease (PD) is a progressive neurodegenerative disorder in ageing populations, yet early diagnosis before motor symptoms remains critical. Reliable identification of subtle nigral alterations at early stages of the disease on magnetic resonance imaging (MRI) remains challenging. This limitation is mainly attributed to the subjective and low sensitivity of manual image interpretation in early PD. Here, we demonstrate a deep learning-based framework to enhance early PD detection. The study’s novelty is a lightweight deep learning framework that captures spatial, textural, and frequency-domain PD biomarkers without heavy network architectures or manual region delineation. Methods: The model integrates GhostNet with ensemble learning to combine local and global spatial information. This model employs wavelet-based frequency feature extraction rather than downsampling and incorporates an attention module to focus on relevant image regions, particularly changes in the substantia nigra (SN) region. Segmentation is employed solely as an auxiliary intermediate step to localize the SN and guide discriminative feature extraction. The final output is a binary classification that distinguishes PD patients from healthy controls. T2-weighted MRI data from the PPMI database are employed. Results: The proposed model achieved an F1-score of 0.8762, demonstrating robust performance under class imbalance, outperforming state-of-the-art models with only 2.03 million parameters and 4.36 Giga Floating Point Operations (GFLOPs). The architecture uncovered texture and frequency patterns previously inaccessible with conventional CNN pipelines. Model comparisons demonstrated consistent gains across all evaluated metrics (all p < 0.001), establishing robust diagnostic improvement. Conclusions: These findings establish an efficient, high-performing framework for reliable MRI-based PD identification. The approach provides automated early detection and supports clinically scalable, computationally lightweight screening tools. Full article

Cotton spider mites pose a significant threat to cotton production, while traditional manual investigation and blanket pesticide application are inefficient for precision pest management in large-scale cotton fields. To address this challenge, this study developed an integrated UAV multispectral remote sensing system for spider mite monitoring and precision spraying. Multispectral imagery was acquired from cotton fields in Shaya County, Xinjiang using UAV-mounted cameras, and vegetation indices including RDVI, MSAVI, SAVI, and OSAVI were selected through feature optimization. Comparative evaluation of three machine learning models (Logistic Regression, Random Forest, and Support Vector Machine) and two deep learning models (1D-CNN and MobileNetV2) was conducted. Considering classification performance and computational efficiency for real-time UAV deployment, Random Forest was identified as optimal, achieving 85.47% accuracy, an 85.24% F1-score, and an AUC of 0.912. The model generated centimeter-level spatial distribution maps for precise spray zone delineation. An improved NSGA-III multi-objective path optimization algorithm was proposed, incorporating PCA-based heuristic initialization, differential evolution operators, and co-evolutionary dual population strategies to optimize deadheading distance, energy consumption, operation time, turning frequency, and load balancing. Ablation study validated the effectiveness of each component, with the fully improved algorithm reducing IGD by 59.94% and increasing HV by 5.90% compared to standard NSGA-III. Field validation showed 98.5% coverage of infested areas with only 3.6% path repetition, effectively minimizing pesticide waste and phytotoxicity risks. This study established a complete technical pipeline from monitoring to application, providing a valuable reference for precision pest control in large-scale cotton production systems. The framework demonstrated robust performance across multiple field sites, though its generalization is currently limited to one geographic region and growth stage. Future work will extend its application to additional cotton varieties, growth stages, and geographic regions. Full article

Giving consideration to cooking activity is important for sustainable housing. In contexts of limited ventilation, imposed by energy saving concerns, cooking causes deterioration of indoor air quality (IAQ) and occupants’ discomfort. This study presents a cooking event detection system that may support IAQ control to minimize the impact of cooking. The system consists of a multi-sensor device and a deep-learning neural network (DNN). The device monitors temperature (T), relative humidity (RH), suspended particulate matter (PM), CO₂, the responses of sensors to volatile organic compounds (VOCs), and other gases (NO₂, CO, CH₂O) in the kitchen zone. The collected data are processed by the DNN. The detection system generates a response every 7 s, indicating either ’COOKING’ or ’NO COOKING’. Feature vector selection was based on classification performance and cost considerations. Cooking event misdetections generate unjustified IAQ control costs: economic ones (UEC), when the system detects a non-existent event, and environmental ones (UEN), when the system fails to detect an actual event. In this study, several well-performing detection systems were developed, with miss rates ranging from 5.1% to 20.5% and false detection rates ranging from 7.7% to 11.7%. The results show that gas sensor responses—particularly to VOCs—had greater utility for cooking event detection compared with T, RH, CO₂, and PM. The cost analysis demonstrated that IAQ control supported by the developed cooking event detection systems could generate higher total unjustified environmental costs when the unit cost ratio UEN/UEC exceeded 1.25, or higher total unjustified economic costs when the unit cost ratio UEN/UEC was below 1.43. We believe this work will contribute to the development of novel automatic IAQ control systems supported by event detection. Full article

Background/Objectives: The precise classification of non-small-cell lung cancer (NSCLC) into lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) has important role in treatment decisions and in prognosis. Proper subtyping ensures that patients receive the most appropriate therapeutic strategies and allows clinicians to make informed evaluations regarding disease outcomes. This study presents a deep neural-network-based classification approach utilizing genome-wide DNA methylation profiles from the Illumina HumanMethylation450 BeadChip platform. Methods: A total of 5000 of the most discriminative CpG probes are identified through variance-based feature selection in the presented methodology, which are then classified through a five-layer deep neural network with batch normalization and dropout regularization. Training and validation were performed using data from The Cancer Genome Atlas (TCGA), with external validation conducted on two independent Gene Expression Omnibus (GEO) datasets: GSE39279 and GSE56044. Results: The model achieved 96.92% accuracy with an area under the receiver-operating characteristic curve (AUC-ROC) of 0.9981 on the TCGA test set. Robust generalization was obtained in cross-dataset validation experiments, with the GEO-trained model achieving 88.92% accuracy and 0.9724 AUC-ROC when validated on TCGA data. The most influential CpG biomarkers contributing to classification decisions are analysed using SHAP (Shapley Additive Explanations). Conclusions: These findings demonstrate the potential of DNA methylation-based deep learning approaches for reliable NSCLC subtype classification with clinical applicability. Full article

Oat is increasingly recognized as a valuable cereal due to its favorable nutritional profile and potential application in functional foods. This study aimed to provide an integrated nutritional and technological evaluation of black oat (Avena strigosa) and white oat (Avena sativa L.) cultivars Ovidiu, Jeremy, and Sorin, grown under uniform conditions. The chemical composition was assessed by determining proteins, lipids, total mineral and polyphenol contents. Macro- and microelement profiles (Ca, Mg, K, Na, Fe, Mn, Cu, Ni, and Zn) were quantified by atomic absorption spectrometry (AAS), while the technological suitability of black oat flour for bakery applications was evaluated using Mixolab analysis and bread quality parameters. Additionally, Fourier-transform infrared (FTIR) spectroscopy was applied to investigate structural features associated with β-glucans in the oat samples. The results showed that protein content ranged from 12.39 to 13.48%, while lipid content varied between 3.24 and 4.64%. Significant differences were observed in mineral composition among the analyzed samples. Black oat showed a balanced mineral profile, characterized by high levels of K, Mg, Mn, Zn, and Ni, confirming its classification as a mineral-rich cereal, while the Ovidiu cultivar generally presented the lowest concentrations for most elements. Mixolab results revealed that the partial substitution of wheat flour with black oat flour significantly influenced dough rheological behavior, particularly in terms of protein weakening and starch gelatinization, without severely affecting dough stability when applied at moderate inclusion levels. Bread quality evaluation demonstrated acceptable crumb elasticity, porosity, and height-to-diameter ratios, supporting the feasibility of incorporating black oat in bakery products. FTIR analysis revealed characteristic absorption bands associated with β-glucans, supporting their presence and structural integrity in both black oat and cultivated varieties. Overall, this study demonstrates that both black oat and selected oat cultivars represent valuable raw materials for functional food applications, offering enhanced nutritional profiles and suitable technological performance. The combined use of compositional, rheological, and spectroscopic analyses provides a comprehensive approach for evaluating oat-based ingredients in the context of modern cereal science. Full article

Rolling bearings are crucial components in CNC machine tool spindles, and their health condition directly affects machining precision and operational reliability. To address the significant challenges of bearing fault diagnosis in industrial environments, this paper proposes an adaptive shapelet-based deep learning model for bearing fault diagnosis. The proposed model integrates three key components: (1) an adaptive multi-scale shapelet extraction module for discriminative pattern learning, (2) a gated parallel CNN with depthwise separable convolutions for multi-scale spatial feature extraction, (3) an enhanced bidirectional long short-term memory network with residual connections for temporal dependency modeling. A composite loss function combining cross-entropy, supervised contrastive learning, and multi-scale consistency regularization is employed for training. To simulate real-world industrial noise conditions, Gaussian, uniform, and impulse noise were injected into the signals. Experiments conducted on the CWRU and IMS datasets demonstrate that, compared with state-of-the-art methods, the proposed approach achieves stronger noise robustness, higher fault classification accuracy, and more stable performance under severe noise contamination. Full article