Search Results (121)

Soluble solid content (SSC) is a key indicator of strawberry quality. Conventional SSC measurement methods are destructive and impractical for large-scale applications. Therefore, this study developed a region-based hyperspectral imaging (HSI) and lightweight one-dimensional convolutional neural network (1D CNN) framework for nondestructive SSC prediction in strawberries. To evaluate spatial effects on predictive accuracy, the fruit surface was segmented into five groups (G1–G5). Three spectral preprocessing methods were applied with partial least squares regression and five convolutional neural network (CNN) architectures, including a simplified VGG-CNN. Larger regions generally improved prediction performance; however, the 50% region (G2) and 75% region (G3) achieved comparable performance to the full region, reducing data requirements. The simplified VGG-CNN model with SNV outperformed other models, exhibiting high accuracy with reduced computational cost, supporting its potential integration into portable and real-time sensing systems. The proposed approach can contribute to improved postharvest quality control and enhanced consumer confidence in strawberry products. Full article

The increasing size of recent deep neural networks, particularly when applied to learning across multiple domains, limits their deployment in resource-constrained environments. To address this issue, this study proposes a lightweight neural architecture with a parallel structure of convolutional layers to enable efficient and scalable multi-domain learning. The proposed network includes an individual feature extractor for domain-specific features and a common feature extractor for the shared features. This design minimizes redundancy and significantly reduces the number of parameters while preserving classification performance. To evaluate the proposed method, experiments were conducted using four image classification datasets: MNIST, FMNIST, CIFAR10, and SVHN. These experiments focused on classification settings where each image contained a single dominant object without relying on large pretrained models. The proposed model achieved high accuracy while significantly reducing the number of parameters. It required only 3.9 M parameters for learning across the four datasets, compared to 33.6 M for VGG16. The model achieved an accuracy of 98.87% on MNIST and 85.83% on SVHN, outperforming other lightweight models, including MobileNet v2 and EfficientNet v2b0, and was comparable to ResNet50. These findings indicate that the proposed architecture has the potential to support multi-domain learning while minimizing model complexity. This approach may be beneficial for applications in resource-constrained environments. Full article

Knowledge distillation achieves model compression by training a lightweight student network to mimic the output distribution of a larger teacher network. However, when the teacher becomes overconfident, its sharply peaked logits break the scale symmetry of supervision and induce high-variance gradients, leading to unstable optimization. Meanwhile, research that focuses only on final-logit alignment often fails to utilize intermediate semantic structure effectively. This causes weak discrimination of student representations, especially under class imbalance. To address these issues, we propose Scale Calibration and Pressure-Driven Knowledge Distillation (SPKD): a one-stage framework comprising two lightweight, complementary mechanisms. First, a dynamic scale calibration module normalizes the teacher’s logits to a consistent magnitude, reducing gradient variance. Secondly, an adaptive pressure-driven mechanism refines student learning by preventing feature collapse and promoting intra-class compactness and inter-class separability. Extensive experiments on CIFAR-100 and ImageNet demonstrate that SPKD achieves superior performance to distillation baselines across various teacher–student combinations. For example, SPKD achieves a score of 74.84% on CIFAR-100 for the homogeneous architecture VGG13-VGG8. Additional evidence from logit norm and gradient variance statistics, as well as representation analyses, proves the fact that SPKD stabilizes optimization while learning more discriminative and well-structured features. Full article

The coronavirus is highly contagious, making rapid early diagnosis essential. Although deep learning-based diagnostic methods using CT or X-ray images have advanced significantly, they still face limitations in cost, processing time, and radiation exposure. In addition, for the possibility of real-time COVID-19 diagnosis, model lightweighting is required. This study proposes a lightweight deep learning model for COVID-19 diagnosis based on fluorescence images and demonstrates its applicability in embedded environments. To prevent data imbalance caused by noise and experimental variations, images were preprocessed using Gray Scale conversion, CLAHE, and Z-Score normalization to equalize brightness values. Among the tested architectures—VGG, ResNet, DenseNet, and EfficientNet—ResNet152 and VGG13 achieved the highest accuracies of 97.25% and 93.58%, respectively, and were selected for lightweighting. Layer-wise importance was calculated using an imprinting-based method, and less important layers were pruned. The pruned VGG13 maintained its accuracy while reducing model size by 18.9 MB and parameters by 4.2 M. ResNet152 (Prune 39) improved accuracy by 1% while reducing size by 161.5 MB and parameters by 40.22 M. The optimized model achieved 129.97 ms, corresponding to 7.69 frames per second (FPS) on an NPU(Furiosa AI Warboy), proving real-time COVID-19 diagnosis is feasible even on low-power edge devices. Full article

Background: Early and accurate cancer detection remains a critical challenge in global healthcare. Deep learning has shown strong diagnostic potential, yet widespread adoption is limited by dependence on high-performance hardware, centralized servers, and data-privacy risks. Methods: This study introduces a browser-based multi-cancer classification framework that performs real-time, client-side inference using TensorFlow.js—eliminating the need for external servers or specialized GPUs. The proposed model fine-tunes the Xception architecture, leveraging depthwise separable convolutions for efficient feature extraction, on a large multi-cancer dataset of over 130,000 histopathological and cytological images spanning 26 cancer types. It was benchmarked against VGG16, ResNet50, EfficientNet-B0, and Vision Transformer. Results: The model achieved a Top-1 accuracy of 99.85% and Top-5 accuracy of 100%, surpassing all comparators while maintaining lightweight computational requirements. Grad-CAM visualizations confirmed that predictions were guided by histopathologically relevant regions, reinforcing interpretability and clinical trust. Conclusions: This work represents the first fully browser-deployable, privacy-preserving deep learning framework for multi-cancer diagnosis, demonstrating that high-accuracy AI can be achieved without infrastructure overhead. It establishes a practical pathway for equitable, cost-effective global deployment of medical AI tools. Full article

This study focuses on the interpretability of diabetic retinopathy classification models. Seven widely used interpretability methods—Gradient, SmoothGrad, Integrated Gradients, SHAP, DeepLIFT, Grad-CAM++, and ScoreCAM—are applied to assess the interpretability of four representative deep learning architectures, VGG, ResNet, DenseNet, and EfficientNet, on fundus images. Through saliency map visualization, perturbation curve analysis, and trend correlation analysis, combined with four quantitative metrics—saliency map entropy, AOPC score, Recall, and Dice coefficient—the interpretability performance of the models is comprehensively assessed from both qualitative and quantitative perspectives. The results show that model architecture greatly influences interpretability quality: models with simpler structures and clearer feature extraction paths (such as VGG) perform better in terms of interpretability, while deeper or lightweight architectures exhibit certain limitations. Full article

Rice, a dietary staple for over half of the global population, is highly susceptible to bacterial and fungal diseases such as bacterial blight, brown spot, and leaf smut, which can severely reduce yields. Traditional manual detection is labor-intensive and often results in delayed intervention and excessive chemical use. Although deep learning models like convolutional neural networks (CNNs) achieve high accuracy, their computational demands hinder deployment in resource-limited agricultural settings. We propose MCP-X, an ultra-compact CNN with only 0.21 million parameters for real-time, on-device rice disease classification. MCP-X integrates a shallow encoder, multi-branch expert routing, a bi-level recurrent simulation encoder–decoder (BRSE), an efficient channel attention (ECA) module, and a lightweight classifier. Trained from scratch, MCP-X achieves 98.93% accuracy on PlantVillage and 96.59% on the Rice Disease Detection Dataset, without external pretraining. Mechanistically, expert routing diversifies feature branches, ECA enhances channel-wise signal relevance, and BRSE captures lesion-scale and texture cues—yielding complementary, stage-wise gains confirmed through ablation studies. Despite slightly higher FLOPs than MobileNetV2, MCP-X prioritizes a minimal memory footprint (~1.01 MB) and deployability over raw speed, running at 53.83 FPS (2.42 GFLOPs) on an RTX A5000. It achieves 16.7×, 287×, 420×, and 659× fewer parameters than MobileNetV2, ResNet152V2, ViT-Base, and VGG-16, respectively. When integrated into a multi-resolution ensemble, MCP-X attains 99.85% accuracy, demonstrating exceptional robustness across controlled and field datasets while maintaining efficiency for real-world agricultural applications. Full article

In recent years, the widespread adoption of photovoltaic (PV) panels for electricity generation has provided significant momentum toward sustainable energy goals. However, it has been observed that the accumulation of dust and contaminants on panel surfaces markedly reduces efficiency by blocking solar radiation from reaching the surface. Consequently, dust detection has become a critical area of research into the energy efficiency of PV systems. This study proposes SolPowNet, a novel Convolutional Neural Network (CNN) model based on deep learning with a lightweight architecture that is capable of reliably distinguishing between images of clean and dusty panels. The performance of the proposed model was evaluated by testing it on a dataset containing images of 502 clean panels and 340 dusty panels and comprehensively comparing it with state-of-the-art CNN-based approaches. The experimental results demonstrate that SolPowNet achieves an accuracy of 98.82%, providing 5.88%, 3.57%, 4.7%, 18.82%, and 0.02% higher accuracy than the AlexNet, VGG16, VGG19, ResNet50, and Inception V3 models, respectively. These experimental results reveal that the proposed architecture exhibits more effective classification performance than other CNN models. In conclusion, SolPowNet, with its low computational cost and lightweight structure, enables integration into embedded and real-time applications. Thus, it offers a practical solution for optimizing maintenance planning in photovoltaic systems, managing panel cleaning intervals based on data, and minimizing energy production losses. Full article

Background and Objectives: Quantifying nasal proportions is central to facial plastic and reconstructive surgery, yet manual measurements are time-consuming and variable. We sought to develop a simple, reproducible deep learning pipeline that localizes the nose in a single frontal photograph and automatically computes the two-dimensional, photograph-derived apparent nasal index (aNI)—width/height × 100—enabling classification into five standard anthropometric categories. Materials and Methods: From CelebA we curated 29,998 high-quality near-frontal images (training 20,998; validation 5999; test 3001). Nose masks were manually annotated with the VGG Image Annotator and rasterized to binary masks. Ground-truth aNI was computed from the mask’s axis-aligned bounding box. A lightweight one-class YOLOv8n detector was trained to localize the nose; predicted aNI was computed from the detected bounding box. Performance was assessed on the held-out test set using detection coverage and mAP, agreement metrics between detector- and mask-based aNI (MAE, RMSE, R²; Bland–Altman), and five-class classification metrics (accuracy, macro-F1). Results: The detector returned at least one accepted nose box in 3000/3001 test images (99.97% coverage). Agreement with ground truth was strong: MAE 3.04 nasal index units (95% CI 2.95–3.14), RMSE 4.05, and R² 0.819. Bland–Altman analysis showed a small negative bias (−0.40, 95% CI −0.54 to −0.26) with limits of agreement −8.30 to 7.50 (95% CIs −8.54 to −8.05 and 7.25 to 7.74). After excluding out-of-range cases (<40.0), five-class classification on n = 2976 images achieved macro-F1 0.705 (95% CI 0.608–0.772) and 80.7% accuracy; errors were predominantly adjacent-class swaps, consistent with the small aNI error. Additional analyses confirmed strong ordinal agreement (weighted κ = 0.71 linear, 0.78 quadratic; Spearman ρ = 0.76) and near-perfect adjacent-class accuracy (0.999); performance remained stable when thresholds were shifted ±2 NI units and across sex and age subgroups. Conclusions: A compact detector can deliver near-universal nose localization and accurate automatic estimation of the nasal index from a single photograph, enabling reliable five-class categorization without manual measurements. The approach is fast, reproducible, and promising as a calibrated decision-support adjunct for surgical planning, outcomes tracking, and large-scale morphometric research. Full article

Image classification is central to computer vision, supporting applications from autonomous driving to medical imaging, yet state-of-the-art convolutional neural networks remain constrained by heavy floating-point operations (FLOPs) and parameter counts on edge devices. To address this accuracy–efficiency trade-off, we propose a unified lightweight framework built on a pruning-aware coordinate attention block (PACB). PACB integrates coordinate attention (CA) with L1-regularized channel pruning, enriching feature representation while enabling structured compression. Applied to MobileNetV3 and RepVGG, the framework achieves substantial efficiency gains. On GTSRB, MobileNetV3 parameters drop from 16.239 M to 9.871 M (–6.37 M) and FLOPs from 11.297 M to 8.552 M (–24.3%), with accuracy improving from 97.09% to 97.37%. For RepVGG, parameters fall from 7.683 M to 7.093 M (–0.59 M) and FLOPs from 31.264 M to 27.918 M (–3.35 M), with only ~0.51% average accuracy loss across CIFAR-10, Fashion-MNIST, and GTSRB. Complexity analysis further confirms PACB does not increase asymptotic order, since the additional CA operations contribute only lightweight lower-order terms. These results demonstrate that coupling CA with structured pruning yields a scalable accuracy–efficiency trade-off under hardware-agnostic metrics, making PACB a promising, deployment-ready solution for mobile and edge applications. Full article

Background/Objectives: Mpox, a viral disease marked by distinctive skin lesions, has emerged as a global health concern, underscoring the need for scalable, accessible, and accurate diagnostic tools to strengthen public health responses. This study introduces ITMA’INN, an AI-driven healthcare system designed to detect Mpox from skin lesion images using advanced deep learning. Methods: The system integrates three key components: an AI model pipeline, a cross-platform mobile application, and a real-time public health dashboard. We leveraged transfer learning on publicly available datasets to evaluate pretrained deep learning models. Results: For binary classification (Mpox vs. non-Mpox), Vision Transformer, MobileViT, Transformer-in-Transformer, and VGG16 achieved peak performance, each with 97.8% accuracy and F1-score. For multiclass classification (Mpox, chickenpox, measles, hand-foot-mouth disease, cowpox, and healthy skin), ResNetViT and ViT Hybrid models attained 92% accuracy (F1-scores: 92.24% and 92.19%, respectively). The lightweight MobileViT was deployed in a mobile app that enables users to analyze skin lesions, track symptoms, and locate nearby healthcare centers via GPS. Complementing this, the dashboard equips health authorities with real-time case monitoring, symptom trend analysis, and intervention guidance. Conclusions: By bridging AI diagnostics with mobile technology and real-time analytics, ITMA’INN advances responsive healthcare infrastructure in smart cities, contributing to the future of proactive public health management. Full article

Spiking Neural Networks (SNNs), with their event-driven and energy-efficient characteristics, have shown great promise in processing data from neuromorphic sensors. However, the sparse and non-stationary nature of event-based data poses significant challenges to optimization, particularly when using conventional algorithms such as AdamW, which assume smooth gradient dynamics. To address this limitation, we propose MemristiveAdamW, a novel algorithm that integrates memristor-inspired dynamic adjustment mechanisms into the AdamW framework. This optimization algorithm introduces three biologically motivated modules: (1) a direction-aware modulation mechanism that adapts the update direction based on gradient change trends; (2) a memristive perturbation model that encodes history-sensitive adjustment inspired by the physical characteristics of memristors; and (3) a memory decay strategy that ensures stable convergence by attenuating perturbations over time. Extensive experiments are conducted on two representative event-based datasets, Prophesee NCARS and GEN1, across three SNN architectures: Spiking VGG-11, Spiking MobileNet-64, and Spiking DenseNet-121. Results demonstrate that MemristiveAdamW consistently improves convergence speed, classification accuracy, and training stability compared to standard AdamW, with the most significant gains observed in shallow or lightweight SNNs. These findings suggest that memristor-inspired optimization offers a biologically plausible and computationally effective paradigm for training SNNs on event-driven data. Full article

Deploying deep learning (DL) models in real-world environments remains a major challenge, particularly under resource-constrained conditions where achieving both high accuracy and compact architectures is essential. While effective, Conventional pruning methods often suffer from high computational overhead, accuracy degradation, or disruption of the end-to-end training process, limiting their practicality for embedded and real-time applications. We present Dynamic Attention-Guided Pruning (DAGP), a Dynamic Attention-Guided Soft Channel Pruning framework that overcomes these limitations by embedding learnable, differentiable pruning masks directly within convolutional neural networks (CNNs). These masks act as implicit attention mechanisms, adaptively suppressing non-informative channels during training. A progressively scheduled L1 regularization, activated after a warm-up phase, enables gradual sparsity while preserving early learning capacity. Unlike prior methods, DAGP is retraining-free, introduces minimal architectural overhead, and supports optional hard pruning for deployment efficiency. Joint optimization of classification and sparsity objectives ensures stable convergence and task-adaptive channel selection. Experiments on CIFAR-10 (VGG16, ResNet56) and PlantVillage (custom CNN) achieve up to 98.82% FLOPs reduction with accuracy gains over baselines. Real-world validation on an enhanced PlantDoc dataset for agricultural monitoring achieves 60 ms inference with only 2.00 MB RAM on a Raspberry Pi 4, confirming efficiency under field conditions. These results illustrate DAGP’s potential to scale beyond agriculture to diverse edge-intelligent systems requiring lightweight, accurate, and deployable models. Full article

The proliferation of smart environments and Internet of Things (IoT) applications has intensified the demand for efficient, privacy-preserving multi-face recognition systems. Conventional centralized systems suffer from latency, scalability, and security vulnerabilities. This paper presents a practical hybrid multi-face recognition framework designed for decentralized IoT deployments. Our approach leverages a pre-trained Convolutional Neural Network (VGG16) for robust feature extraction and a Support Vector Machine (SVM) for lightweight classification, enabling real-time recognition on resource-constrained devices such as IoT cameras and Raspberry Pi boards. The purpose of this work is to demonstrate the feasibility and effectiveness of a lightweight hybrid system for decentralized multi-face recognition, specifically tailored to the constraints and requirements of IoT applications. The system is validated on a custom dataset of 20 subjects collected under varied lighting conditions and facial expressions, achieving an average accuracy exceeding 95% while simultaneously recognizing multiple faces. Experimental results demonstrate the system’s potential for real-world applications in surveillance, access control, and smart home environments. The proposed architecture minimizes computational load, reduces dependency on centralized servers, and enhances privacy, offering a promising step toward scalable edge AI solutions. Full article

Activation functions play a crucial role in ensuring training stability, convergence speed, and overall performance in both convolutional and attention-based networks. In this study, we introduce two novel activation functions, each incorporating a sine component and a constraint term. To assess their effectiveness, we replace the activation functions in four representative architectures—VGG16, ResNet50, DenseNet121, and Vision Transformers—covering a spectrum from lightweight to high-capacity models. We conduct extensive evaluations on four benchmark datasets (CIFAR-10, CIFAR-100, MNIST, and Fashion-MNIST), comparing our methods against seven widely used activation functions. The results consistently demonstrate that our proposed functions achieve superior performance across all tested models and datasets. From a design application perspective, the proposed functional periodic structure also facilitates rich and structurally stable activation visualizations, enabling designers to trace model attention, detect surface biases early, and make informed aesthetic or accessibility decisions during interface prototyping. Full article