Search Results (218)

Human sitting posture recognition is critical for smart seating, ergonomic monitoring, and healthcare systems. However, existing deep learning approaches typically rely on highly complex network architectures that are computationally expensive, hindering their lightweight deployment on edge devices. Furthermore, current methods frequently struggle with indistinct boundaries among multi-class postures and are highly prone to overfitting when constrained by small-sample pressure sensor datasets. To bridge this gap, this paper proposes a novel, lightweight posture recognition framework specifically tailored for pressure distribution maps. First, sitting pressure data is collected using a thin-film pressure array sensor and uniformly mapped into an [M × N] image representation, establishing an effective sample format for Convolutional Neural Network (CNN) inputs. Second, as our primary architectural contribution, we fundamentally optimize the classic LeNet-5 network to enhance complex feature representation without inflating model complexity. Specifically, the depth of the convolutional layers is increased with a progressively increasing channel configuration. Batch Normalization (BN) is introduced to accelerate convergence and ensure training stability, while a Dropout mechanism is embedded within the fully connected layers to strictly penalize overfitting under small-sample constraints. These architectural improvements are synergistically combined with targeted data augmentation strategies—including random translation, rotation, and intensity perturbation—to further strengthen the model’s generalization capability. Experimental results demonstrate that the proposed method achieves a classification accuracy of 95.5% in a five-class sitting posture recognition task, significantly outperforming baseline models such as the traditional LeNet-5, AlexNet-Lite, and VGG-Small. The findings indicate that this approach achieves an optimal balance among recognition accuracy, training stability, and low model complexity, providing a robust algorithmic baseline and proof-of-concept for smart healthcare perception systems, paving the way for future large-scale subject-independent validation. Full article

Deep learning has significantly advanced image super-resolution (SR), yet many state-of-the-art models remain too computationally expensive for resource-constrained devices. This paper demonstrates that a highly parameter-efficient design can achieve comparable performance to the very deep super-resolution network (VDSR) with a tiny fraction of parameters. Starting from the classic VDSR architecture (2016), we systematically evaluate three design choices: depthwise separable convolution (DSConv), Hybrid Attention Transformer (HAT), and a local residual connection (LR). HAT provides no performance gain—an honest negative result supported by controlled experiments (increased training, different reduction ratios, and standard convolution baseline). In contrast, LR alone yields a 0.20 dB improvement without introducing any extra parameters. Consequently, we discard HAT and propose DSConv+LR. Our model contains only 49,217 parameters—about 7.4% of VDSR—yet attains a peak signal-to-noise ratio (PSNR) of 35.21 dB on Set5 (×2), which is 99.7% of VDSR’s performance (35.33 dB). On additional benchmarks (Set14, BSD100, and Urban100), DSConv+LR maintains similar relative performance (within 0.12 dB of VDSR). Perceptual loss (AlexNet features, lower better) is 0.2556, slightly better than VDSR (0.2717). We acknowledge that modern lightweight networks such as cascaded residual attention network (CARN) and information multi-distillation network (IMDN) achieve 2–3 dB higher PSNR at the cost of 9–14× more parameters. This work advocates a minimalist approach while honestly reporting both its strengths and limitations. Full article

Integrating complementary attention mechanisms into standard Convolutional Neural Networks (CNNs) is a promising strategy for improving feature discrimination without substantial computational overhead. This paper presents a controlled empirical study of a hybrid attention module that combines Squeeze-and-Excitation Networks (SENet) and the Convolutional Block Attention Module (CBAM) through an adaptive element-wise summation with a learnable weighting parameter $\alpha$ and a residual connection. This work contributes a systematic and statistically rigorous evaluation of attention fusion across four CNN backbones (ResNet18, VGG16, AlexNet, and SqueezeNet) on the CIFAR-10 benchmark at $32\times 32$ resolution. All models were trained from scratch under a deliberately conservative protocol (50 epochs, no pretrained weights, standard augmentation) to isolate the incremental effect of attention mechanisms under controlled conditions. Under this protocol, the hybrid SENet+CBAM configuration achieves statistically significant accuracy improvements over the corresponding baselines ($p<0.001$, 5-fold cross-validation): ResNet18 improves from 77.93% to 90.71% (+12.78%), VGG16 from 55.78% to 70.17% (+14.39%), AlexNet from 62.67% to 71.82% (+9.15%), and SqueezeNet from 71.91% to 78.29% (+6.38%). These gains must be interpreted within the scope of this controlled setting. Absolute accuracy values are below fully optimized literature benchmarks. For VGG16 in particular, part of the improvement likely reflects correction of underfitting under the conservative protocol, not the full potential of the hybrid mechanism. Parameter overhead remains modest at 1.5–5.8%, and training convergence improves by 16.5% on average. The hybrid approach outperforms the best previously reported SENet+CBAM result for each architecture by an average of 2.32%. Grad-CAM visualizations and attention entropy analysis provide qualitative evidence of more concentrated spatial attention patterns under the hybrid configuration. These should be understood as proxy indicators rather than rigorous interpretability measures. Validation on higher-resolution benchmarks such as CIFAR-100, STL-10, and ImageNet subsets is a necessary next step before broader applicability can be claimed. Full article

The precise and timely diagnosis of grapevine diseases is paramount for ensuring food security and mitigating economic losses within the viticulture sector. While existing deep learning models offer high accuracy, their computational intensity and hardware requirements often hinder their use in portable or low-power field systems. This study addresses this gap by proposing GrapeLeafNet, a lightweight convolutional neural network optimized for efficient feature extraction. GrapeLeafNet introduces a strategic hybrid approach that combines the low parameter efficiency of models like SqueezeNet with the rapid feature propagation advantages offered by shallow architectures such as AlexNet. By eliminating the sequential processing latency caused by SqueezeNet’s 18-layer deep structure and the excessive 61-million-parameter memory burden of AlexNet, this model establishes a critical balance between low latency and high accuracy through its optimized 7-layer architecture. Characterized by an original integration of standard convolutional layers, batch normalization, and max pooling, GrapeLeafNet achieves high computational efficiency with only 1.6 million parameters and a 6.26 MB memory footprint. This structural optimization enhances deep feature hierarchies, enabling the model to focus on distinctive pathological signs within complex leaf patterns and maximize classification sensitivity by filtering out irrelevant features. The evaluation was conducted using the Niphad Grape Leaf Disease (NGLD) dataset, incorporating data augmentation to mitigate inherent class imbalances. Additionally, data augmentation techniques were employed to mitigate inherent class imbalances within the dataset. Experimental results demonstrate that GrapeLeafNet achieved 97.06% accuracy and a 94.77% F1-score on the original dataset, outperforming recent benchmarks by 2.46%. Following augmentation, performance reached 98.29% accuracy and a 98.16% F1-score, representing a 6.16% higher F1-score than contemporary models. GrapeLeafNet exhibits high robustness against asymmetric class distributions and establishes a significant performance margin over existing architectures. Its lightweight nature, combined with superior accuracy and F1-score metrics, makes it an ideal candidate for integration into mobile devices and real-time agricultural monitoring systems. Full article

This paper investigates the design and evaluation of compact convolutional neural networks (CNNs) for keyword spotting (KWS) and acoustic event detection under the stringent constraints of the TinyML paradigm. The research expands upon traditional binary classification approaches by addressing a multi-class acoustic scenario encompassing eight distinct categories: stop, no, go, yes, unknown, silence, noise_ambient, and noise_sudden. The primary objective is to evaluate the feasibility of deploying reliable acoustic detection systems on ultra-low-power microcontrollers for edge computing applications. To this end, five lightweight architectures were developed and benchmarked: AlexNet-Tiny, LeNet-Tiny, MobileNet-Tiny, VGG-Tiny, and CustomCNN-Tiny. The models were trained using Mel-spectrogram features and optimized through INT8 post-training quantization to facilitate embedded deployment. Hardware simulation was conducted targeting the XIAO nRF52840 Sense microcontroller (64 MHz, 256 KB RAM). Experimental results demonstrate that the Gold VGG-Tiny architecture achieves the highest classification accuracy ($89.81\%$), while Silver MobileNet-Tiny provides the superior operational efficiency with the lowest inference latency ($0.88$ ms) and minimal energy consumption ($14.4$ µJ). Furthermore, the Bronze CustomCNN-Tiny model achieves the most reduced memory footprint ($42.9$ KB), highlighting its suitability for memory-constrained environments. Statistical validation using Cohen’s Kappa, Matthews Correlation Coefficient (MCC), and Area Under the Curve (AUC) confirms the robustness and reliability of the proposed models. The potential application of this system is motivated by acoustic monitoring for the early detection of high-risk situations, such as gender-based violence. Future work will focus on on-device physical validation and real-world deployment in wearable safety electronics. Full article

The accelerating growth of scientific publications has intensified the need for scalable and interoperable tools capable of supporting bibliometric analysis and research evaluation. In response to this challenge, this paper introduces IllustryFlow, a modular framework that combines n8n, an open-source workflow automation engine, with Illustry, a dynamic visualization platform, to extract, classify, and interpret scholarly data retrieved from OpenAlex. At the core of the framework is a multilingual BERT-based classification model implemented within the OpenAlex infrastructure, trained on the CWTS (Centre for Science and Technology Studies from Leiden University) classification schema and enriched with metadata features such as journal-level embeddings and citation graph information. IllustryFlow enables automated topic classification, clustering, and semantic visualization of citation networks, co-authorship structures, and thematic distributions. In this framework, Illustry and the custom n8n nodes represent components developed by the author, while OpenAlex and the OpenAlex-enhanced BERT model are integrated as external resources. The principal contribution of this study therefore consists of the architectural design and operational integration of these components into a unified, modular, automated, and reproducible bibliometric workflow. The proposed framework integrates an explicit and reproducible strategy for querying, semantic filtering, and selection of the bibliographic corpus. The framework was evaluated on a dataset of 1756 bibliographic records, and the entire workflow, including dashboard generation, was completed in approximately 90 s under the experimental conditions considered. The obtained results support the feasibility of the framework for scalable bibliometric workflows and indicate its practical potential for the analysis of heterogeneous bibliographic corpora while maintaining reproducibility under the analyzed conditions. Full article

Accurate fruit ripeness assessment is essential for optimizing harvest timing and ensuring postharvest quality, particularly in climacteric fruits such as peaches, which exhibit rapid and heterogeneous ripening. This study proposes a deep learning-based approach for multistage peach ripeness classification across five maturity levels (green, green-blushed, blushed, yellow-blushed, and fully yellow). Four datasets were constructed using controlled image acquisition, segmentation, data augmentation, and perceptual hashing to mitigate data leakage. The performance of AlexNet, EfficientNet-B0, and three YOLO (You Only Look Once) architectures (YOLOv8, YOLOv11, and YOLOv12) was evaluated using standard metrics, including accuracy, precision, recall, F1 score, mAP, and inference speed. Results show that YOLO-based models significantly outperform classical networks, achieving accuracies between 95.25% and 98.3% and mAP@0.5 above 98.25%, while also reducing inference time to 8.1–12.7 ms compared with 722.23 ms for AlexNet and 171.87 ms for EfficientNet-B0. In a practical sorting experiment with 214 peaches, YOLOv12 achieved 92.06% accuracy, demonstrating robust real-world performance. Misclassifications were primarily observed between adjacent ripeness stages. These findings indicate that YOLO-based models provide an effective and scalable solution for real-time fruit sorting, while the use of perceptual hashing enhances dataset reliability and model generalization for deployment in agricultural quality control systems. Full article

The rapid expansion of English major enrollment has exposed critical limitations in traditional academic assessment methods regarding efficiency and accuracy, constraining educational quality enhancement. This paper introduces an English proficiency assessment approach utilizing an improved RegNet architecture integrated with a dual attention mechanism. The multidimensional academic data processed by our model include attendance, online participation, language practice, and assessment scores for listening, speaking, reading, and writing from undergraduate English majors. The initial downsampling module of RegNet is optimized through a dual convolutional structure to augment shallow feature extraction. Subsequently, a deformable attention mechanism (DAT) is incorporated to enhance focus on salient features, while a graph attention network (GAT) facilitates interaction and fusion among academic node features. Experimental results demonstrate that the proposed method achieves an average accuracy of 99.46% in proficiency assessment, substantially outperforming mainstream models including EfficientNet and AlexNet. Additionally, it demonstrates robust edge deployment capabilities, providing an effective technical solution for intelligent academic management of English programs within smart campus frameworks. Full article

Breast cancer histopathology classification remains a fundamental challenge in computational pathology due to variations in tissue morphology across magnification levels. Convolutional neural networks (CNNs) have long been the standard for image-based diagnosis, yet recent advances in vision-language models (VLMs) suggest they may provide strong and transferable representations for complex medical images. In this study, we present a systematic comparison between CNN baselines and large VLMs—Qwen2 and SmolVLM—fine-tuned with Low-Rank Adaptation (LoRA; $r=16$, $\alpha =32$, dropout = $0.05$) on the BreakHis dataset. Models were evaluated at $40\times$, $100\times$, $200\times$, and $400\times$ magnifications using accuracy, precision, recall, F1-score, and area under the ROC curve (AUC). While Qwen2 achieved moderate performance across magnifications (e.g., 0.8736 accuracy and 0.9552 AUC at $200\times$), SmolVLM consistently outperformed Qwen2 and substantially reduced the gap with CNN baselines, reaching up to 0.9453 accuracy and 0.9572 F1-score at $200\times$—approaching the performance of AlexNet (0.9543 accuracy) at the same magnification. CNN baselines, particularly ResNet34, remained the strongest models overall, achieving the highest performance across all magnifications (e.g., 0.9879 accuracy and 0.9984 AUC at $40\times$). These findings demonstrate that LoRA fine-tuned VLMs, despite requiring gradient accumulation and memory-efficient optimizers and operating with a significantly smaller number of trainable parameters, can achieve competitive performance relative to traditional CNNs. However, CNN-based architectures still provide the highest accuracy and robustness for histopathology classification. Our results highlight the potential of VLMs as parameter-efficient alternatives for digital pathology tasks, particularly in resource-constrained settings. Full article

Biometric systems are widely used in security and forensic applications. Conventionally, contact-based footprint scanners require physical contact, which presents significant limitations. These devices raise hygiene concerns and are impractical in field identification conditions, such as forensic investigations or disaster victim identification, where quick and non-invasive methods are essential. To address these challenges, a contactless footprint acquisition and identification system was developed using image processing techniques and a Convolutional Neural Network (CNN) based on the Visual Geometry Group–16 layer architecture. The system employs a Raspberry Pi 4, a Logitech C922 camera, and a ring light to capture footprints without direct surface contact. Captured images are processed with Contrast Limited Adaptive Histogram Equalization (CLAHE) to improve contrast and mean thresholding to generate binary images for clearer feature extraction. System performance was evaluated using a multiclass confusion matrix. The CNN correctly classified 158 of 160 test images, achieving an accuracy of 98.75%. This result demonstrates higher accuracy than earlier studies that used older CNN models, such as Alex Krizhevsky’s Network and LeCun’s Network-5, which performed with fewer subjects and lower accuracy rates. The developed system shows potential for biometric security, forensic investigations, and disaster response, where contactless and reliable identification is required. Future research can expand the dataset with more diverse footprints, test performance under varied conditions, and extend the approach to other contactless biometrics such as palmprints or ears. Full article

Cybersecurity operations in IoT and edge environments require fast, evidence-grounded decisions under strict resource and trust constraints. While large language models can support triage and incident analysis, their parametric knowledge may be outdated and prone to hallucination. Retrieval-augmented generation (RAG) improves grounding by conditioning responses on retrieved evidence, but also introduces new risks such as knowledge-base poisoning, indirect prompt injection, and embedding leakage. Federated learning enables collaborative adaptation without centralizing sensitive data, motivating federated RAG (FedRAG) architectures for distributed cybersecurity deployments. This study presents a deployment-oriented scoping review of FedRAG for cybersecurity. The review follows PRISMA-ScR reporting guidance and synthesizes 82 studies published between 2020 and 2026, identified through keyword search and citation snowballing over OpenAlex, arXiv, and Crossref. We develop a taxonomy that clarifies the components of federated systems, deployment locations, trust boundaries, and protected assets. We further map the combined RAG+FL attack surface, summarize practical defenses and system patterns, and distill actionable guidance for secure, privacy-preserving, and efficient FedRAG deployment in real-world IoT and edge scenarios. Our synthesis highlights recurring trade-offs among robustness, privacy, latency, communication overhead, and maintainability, and identifies open research priorities in benchmark design, governance mechanisms, and cross-silo evaluation protocols for practical deployment. Full article

The main objective of this study is to automatically detect real-time snow-related road surface conditions using imagery captured from existing roadside webcams along interstate freeways. Blowing snow is considered one of the most hazardous roadway weather phenomena because it significantly reduces driver visibility and adversely affects vehicle operation. A comprehensive image preprocessing and reduction process was conducted to construct two reference datasets. The first dataset consisted of two categories (blowing snow and no blowing snow), while the second dataset included five surface condition categories: blowing snow, dry, slushy, snow covered, and snow patched. Eight pre-trained convolutional neural networks (CNNs), including AlexNet, SqueezeNet, ShuffleNet, ResNet18, GoogleNet, ResNet50, MobileNet-V3, and EfficientNet-B0, were evaluated for roadway surface condition classification. For Dataset 1, ResNet50 achieved the highest detection accuracy of 97.88%, while AlexNet demonstrated competitive performance with 97.56% accuracy and significantly shorter training time. Among the lightweight architectures, MobileNet-V3 achieved 95.56% accuracy, demonstrating strong computational efficiency. EfficientNet-B0 achieved 93.56% accuracy while maintaining reduced model complexity. For Dataset 2, ResNet18 achieved the highest multi-class detection accuracy of 96.10%, while AlexNet required the shortest training time among the evaluated models. A comparative analysis between deep CNN models and traditional machine learning approaches showed that deep CNNs significantly outperform feature-based methods in detecting blowing snow conditions. The proposed framework provides an automated, accurate, and scalable solution for roadway surface condition monitoring and supports real-time applications in intelligent transportation systems. Full article

Bone scintigraphy is a widely available and cost-effective modality for detecting skeletal metastases in prostate cancer, yet visual interpretation can be challenging due to heterogeneous uptake patterns, benign mimickers, and a high reporting workload, motivating robust computer-aided decision support. In this study, we present an experimental evaluation of fourteen convolutional neural network (CNN) architectures for binary metastasis classification in planar bone scintigraphy using a unified protocol. Fourteen models, CNN (baseline), AlexNet, VGG16, VGG19, ResNet18, ResNet34, ResNet50, ResNet50-attention, DenseNet121, DenseNet169, DenseNet121-attention, WideResNet50_2, EfficientNet-B0, and ConvNeXt-Tiny, were trained and tested on 600 scan images (300 normal, 300 metastatic) from the Jordanian Royal Medical Services under identical preprocessing and augmentation with stratified five-fold cross-validation. We report mean ± SD for AUC-ROC, accuracy, precision, sensitivity (recall), F1-score, specificity, and Cohen’s κ, alongside calibration via the Brier score and deployment indicators (parameters, FLOPs, model size, and inference time). DenseNet121 achieved the best overall balance of diagnostic performance and reliability, reaching AUC-ROC 96.0 ± 1.2, accuracy 89.2 ± 2.2, sensitivity 83.7 ± 3.4, specificity 94.7 ± 2.2, F1-score 88.5 ± 2.5, κ = 0.783 ± 0.045, and the strongest calibration (Brier 0.080 ± 0.013), with stable fold-to-fold behaviour. DenseNet121-attention produced the highest AUC-ROC (96.3 ± 1.1) but exhibited greater variability in specificity, indicating less consistent false-alarm control. Complexity analysis supported DenseNet121 as deployable (~7.0 M parameters, ~26.9 MB, ~92 ms/image), whereas heavier models yielded only limited additional clinical value. These results support DenseNet121 as a reliable backbone for automated metastasis detection in planar scintigraphy, with future work focusing on external validation, threshold optimisation, interpretability, and model compression for clinical adoption. Full article

Strong gravitational lensing provides valuable insights into the mass distribution of galaxies and the nature of dark matter. However, its modeling is computationally demanding due to the large volume of strong lensing observations. In this work, we explore the application of Convolutional Neural Networks to infer physical parameters from simulated galaxy–galaxy lens systems, described by the Singular Isothermal Ellipsoid (SIE) profile for the galaxy lens. We construct a dataset of 76,396 synthetic lensing images derived from the China Space Station Telescope catalog and employ it to train a modified CNN model, based on the AlexNet architecture, to predict four key SIE parameters, the Einstein radius, the axis ratio and ellipticity components. We analyze the network performance under three distinct dropout configurations to quantify their influence on generalization and parameter inference accuracy. The results indicate that the incorporation of dropout is critical for enhancing the precision and robustness of the estimated parameters as demonstrated using a 4-fold cross-validation procedure. When dropout tools are included, we obtain coefficients of determination up to $R^2\sim 0.96$ for most SIE parameters and mean peak signal-to-noise ratios of up to ∼37 dB. Relative to the configuration without dropout, the use of dropout reduces the relative errors in the inferred SIE parameters by approximately 60–76%, resulting in errors of at most ∼9% at the 90% confidence level for the majority of parameters. These findings highlight the potential of deep learning approaches to enable scalable, computationally efficient, and high-precision modeling of strong gravitational lensing systems. Full article

Background/Objectives: Musculoskeletal disorders and their surgical treatment significantly affect global disability, healthcare utilization, and costs. Artificial intelligence (AI) is a key enabler of data-driven musculoskeletal care. Their applications include diagnostic imaging, surgical planning, risk prediction, rehabilitation, and digital health ecosystems. This narrative review synthesizes current evidence on the use of AI in orthopaedics and musculoskeletal care across five areas: diagnostic imaging, surgical planning and intraoperative augmentation, predictive analytics and patient-reported outcomes, rehabilitation intelligence and teleorthopaedics, and system-level management. An additional task is to identify translational gaps and priorities for safe, ethical, and equitable implementation of AI. Methods: A structured narrative review was conducted using targeted searches in PubMed, Scopus, and Web of Science supplemented by semantic and citation-based explorations in Semantic Scholar, OpenAlex, and Google Scholar. The main search period was January 2019 to December 2025. The retrieved peer-reviewed articles were analyzed for clinical relevance to human musculoskeletal care, quantitative outcomes, and the translational implications of the results. From the broader pool of eligible publications, 40 clinically relevant studies were selected for detailed synthesis covering imaging, surgical planning, predictive modeling, rehabilitation, and system-level applications. Owing to the significant heterogeneity in the model architectures, datasets, and endpoints, the results were organized into five predefined thematic areas. Results: The most mature evidence is for AI-assisted detection of bone fractures on radiographs, identification of implants, and use of sizing templates in preoperative planning for arthroplasty, where deep learning systems have achieved expert-level diagnostic performance (e.g., fracture detection sensitivity of approximately 90% and specificity of approximately 92% and implant identification accuracy of 97–99%) and improved the accuracy of preoperative planning compared to conventional templating. AI-based planning increases the likelihood of reducing intraoperative corrections, shortening surgery time, reducing blood loss, and improving the final functional outcomes. Predictive models can support the stratification of risk for complications, rehospitalizations, and patient-reported outcomes, although external validation remains limited and is often single-center at this stage of research. Emerging applications in rehabilitation and teleorthopaedics, including sensor-based monitoring and learning systems integrated with Patient-Reported Outcome Measures (PROMs), are conceptually promising, but are mainly limited to feasibility or pilot studies. Conclusions: AI is beginning to influence musculoskeletal care, moving beyond pattern recognition toward integrated, patient-centered decision support throughout the perioperative and rehabilitation periods. Its widespread use remains constrained by limited multicenter validation, dataset bias, algorithmic opacity, and immature regulatory and governance frameworks. Future work should prioritize prospective multicenter impact studies, repeatable revalidation of local models, integration of PROM and teleorthopedic data with health learning systems, and adaptation to changing regulatory requirements to enable safe, ethical, effective, and equitable implementation in routine orthopedic practice. Full article