Search Results (7,354)

Accurate cerebral vasculature segmentation using digital subtraction angiography (DSA) is critical for diagnosing and treating cerebrovascular diseases. However, conventional single-frame analysis methods often fail to capture fine vascular structures due to background noise, overlapping anatomy, and dynamic contrast flow. In this study, we propose a novel vessel-enhancing preprocessing technique using temporal differencing of DSA sequences to improve cerebrovascular segmentation accuracy. Our method emphasizes contrast flow dynamics while suppressing static background components by computing absolute differences between sequential DSA frames. The enhanced images were input into state-of-the-art deep learning models, U-Net++ and DeepLabv3+, for vascular segmentation. Quantitative evaluation of the publicly available DIAS dataset demonstrated significant segmentation improvements across multiple metrics, including the Dice Similarity Coefficient (DSC), Intersection over Union (IoU), and Vascular Connectivity (VC). Particularly, DeepLabv3+ with the proposed preprocessing achieved a DSC of 0.83 ± 0.05 and VC of 44.65 ± 0.63, outperforming conventional methods. These results suggest that leveraging temporal information via input enhancement substantially improves small and complex vascular structure extraction. Our approach is computationally efficient, model-agnostic, and clinically applicable for DSA. Full article

This paper presents a novel large-span ring deployable perimeter truss for the mesh reflector deployable antennas, which is made up of two parts including a single-mobility driving mechanism and a ring deployable metamorphic mechanism. The mechanism design employs polygon approximation, and each side is treated as a basic unit using a modular design approach. By reasonable assembly, a ring deployable metamorphic mechanism with a small folded state and a large deployed state can be formed. Here, multiple singular positions, the axis of its three revolute joints being parallel and coplanar, are used in the fully deployed state, which forms multiple dead-center positions and changes the constraint conditions. The metamorphic motion is thus achieved, and a stable self-locking state is established that greatly enhances the stability. The paper first introduces the mechanism design and evaluation method; the kinematic and dynamic analysis is then conducted, and the simulation validation is also performed. Moreover, a principle design for cable-net structural setting and connection is illustrated. Finally, with the design of a driving system and the fabrication of a physical prototype, the deployable experiments are carried out, and the results show that the perimeter truss can efficiently act as the mesh reflector deployable antennas. Full article

Downhole torsional vibrations, especially high-frequency torsional oscillations (HFTOs) and stick–slip phenomena, pose a serious threat to drilling operations, often resulting in tool damage, prolonged non-productive time, and significant cost increases. Traditional monitoring methods cannot promptly capture complex vibration patterns, so there is an urgent need for advanced early warning systems. This study proposes the DS-DW-TimesNet model, which improves the TimesNet framework by incorporating downsampling technology for efficient data compression, dilated convolution that can expand the temporal receptive field, and a learnable weight normalization method that can stabilize the training process, thereby enhancing the capabilities of feature extraction and long-sequence modeling. Verified using field data from the Fuman Oilfield, the results show that in terms of the mean absolute error (MAE) for 210 s predictions, this model is 77.2% and 21.8% lower than LSTM and Informer, respectively, and the inference speed is increased by 78.5% (reaching 48 milliseconds). It can provide reliable 210 s early warning windows for high-frequency torsional oscillations and 150 s early warning windows for stick–slip, exceeding industry standards and helping to improve the safety and efficiency of drilling operations. Full article

Efficient and explainable vehicle damage inspection is essential due to the increasing complexity and volume of vehicular incidents. Traditional manual inspection approaches are not time-effective, prone to human error, and lead to inefficiencies in insurance claims and repair workflows. Existing deep learning methods, such as CNNs, often struggle with generalization, require large annotated datasets, and lack interpretability. This study presents a robust and interpretable deep learning framework for vehicle damage classification, integrating Vision Transformers (ViTs) and ensemble detection strategies. The proposed architecture employs a RetinaNet backbone with a ViT-enhanced detection head, implemented in PyTorch using the Detectron2 object detection technique. It is pretrained on COCO weights and fine-tuned through focal loss and aggressive augmentation techniques to improve generalization under real-world damage variability. The proposed system applies the Weighted Box Fusion (WBF) ensemble strategy to refine detection outputs from multiple models, offering improved spatial precision. To ensure interpretability and transparency, we adopt numerous explainability techniques—Grad-CAM, Grad-CAM++, and SHAP—offering semantic and visual insights into model decisions. A custom vehicle damage dataset with 4500 images has been built, consisting of approximately 60% curated images collected through targeted web scraping and crawling covering various damage types (such as bumper dents, panel scratches, and frontal impacts), along with 40% COCO dataset images to support model generalization. Comparative evaluations show that Hybrid ViT-RetinaNet achieves superior performance with an F1-score of 84.6%, mAP of 87.2%, and 22 FPS inference speed. In an ablation analysis, WBF, augmentation, transfer learning, and focal loss significantly improve performance, with focal loss increasing F1 by 6.3% for underrepresented classes and COCO pretraining boosting mAP by 8.7%. Additional architectural comparisons demonstrate that our full hybrid configuration not only maintains competitive accuracy but also achieves up to 150 FPS, making it well suited for real-time use cases. Robustness tests under challenging conditions, including real-world visual disturbances (smoke, fire, motion blur, varying lighting, and occlusions) and artificial noise (Gaussian; salt-and-pepper), confirm the model’s generalization ability. This work contributes a scalable, explainable, and high-performance solution for real-world vehicle damage diagnostics. Full article

Underwater biomimetic robotic fish are emerging as vital platforms for ocean exploration tasks such as environmental monitoring, biological observation, and seabed investigation, particularly in areas inaccessible to humans. Central to their effectiveness is high-precision fish pose estimation, which enables detailed analysis of swimming patterns and ecological behavior, while informing the design of agile, efficient bio-inspired robots. To address the widespread scarcity of high-quality motion datasets in this domain, this study presents a custom-built dual-camera experimental platform that captures multi-view sequences of carp exhibiting three representative swimming behaviors—straight swimming, backward swimming, and turning—resulting in a richly annotated dataset. To overcome key limitations in existing pose estimation methods, including heavy reliance on labeled data and inadequate modeling of temporal dependencies, a novel Semi-supervised Temporal Context-Aware Network (STC-Net) is proposed. STC-Net incorporates two innovative unsupervised loss functions—temporal continuity loss and pose plausibility loss—to leverage both annotated and unannotated video frames, and integrates a Bi-directional Convolutional Recurrent Neural Network to model spatio-temporal correlations across adjacent frames. These enhancements are architecturally compatible and computationally efficient, preserving end-to-end trainability. Experimental results on the proposed dataset demonstrate that STC-Net achieves a keypoint detection RMSE of 9.71, providing a robust and scalable solution for biological pose estimation under complex motion scenarios. Full article

This research investigates the glycerol oxidation reaction on carbon-functionalized composites using Tafel behavior, exchange current density (ECD), rate constant, and dynamic electrochemical impedance spectroscopy (DEIS) data fitting. The aim is to gather essential data for fabricating glycerol electrooxidation electrodes in an alkaline medium. Corrected net current analysis reveals that the nitrogen-doped activated carbon black composite electrode (ACB-N2) exhibits the highest instantaneous catalytic activity, with a net current density of 1.3 mA cm⁻² at 1.0 V vs. SCE. However, the dual-doped nitrogen and fluorine composite (ACB-N2F2) demonstrates the lowest Tafel slope (177.97 mV dec⁻¹), indicating faster kinetics, and it maintains superior electrochemical stability during chronoamperometric testing. ACB-N2F2 exhibits the highest ECD (1.0129 mA cm⁻²) and the lowest Ts and rate constant (2.62 × 10⁹ cm s⁻¹), indicating the fastest electron transfer. These findings suggest that while ACB-N2 offers the highest net GOR activity, ACB-N2F2 combines kinetic efficiency and long-term durability, making it a promising candidate for practical GOR applications. The rate-determining step is water adsorption at low overpotentials (0.55 V, 0.5 V, and 0.4 V vs. SCE for ACB-F2, ACB-N2, and ACB-N2F2, respectively). Full article

This study developed a deep learning–based water level recognition model using Closed-Circuit Television (CCTV) footage. The model focuses on real-time water level recognition in agricultural reservoirs that lack automated water level gauges, with the potential for future extension to flood forecasting applications. Video data collected over approximately two years at the Myeonggyeong Reservoir in Chungcheongbuk-do, South Korea, were utilized. A semantic segmentation approach using the U-Net model was employed to extract water surface areas, followed by the classification of water levels using Convolutional Neural Network (CNN), ResNet, and EfficientNet models. To improve learning efficiency, water level intervals were defined using both equal spacing and the Jenks natural breaks classification method. Among the models, EfficientNet achieved the highest performance with an accuracy of approximately 99%, while ResNet also demonstrated stable learning outcomes. In contrast, CNN showed faster initial convergence but lower accuracy in classifying complex intervals. This study confirms the feasibility of applying vision-based water level prediction technology to flood-prone agricultural reservoirs. Future work will focus on enhancing system performance through low-light video correction, multi-sensor integration, and model optimization using AutoML, thereby contributing to the development of an intelligent, flood-resilient water resource management system. Full article

Diabetic retinopathy (DR), a critical ocular disease that can lead to blindness, demands early and accurate diagnosis to prevent vision loss. Current automated DR diagnosis methods face two core challenges: first, subtle early lesions such as microaneurysms are often missed due to insufficient feature extraction; second, there is a persistent trade-off between model accuracy and efficiency—lightweight architectures often sacrifice precision for real-time performance, while high-accuracy models are computationally expensive and difficult to deploy on resource-constrained edge devices. To address these issues, this study presents a novel deep learning framework integrating depthwise separable convolution and a multi-view attention mechanism (MVAM) for efficient DR diagnosis using retinal images. The framework employs multi-scale feature fusion via parallel 3 × 3 and 5 × 5 convolutions to capture lesions of varying sizes and incorporates Gabor filters to enhance vascular texture and directional lesion modeling, improving sensitivity to early structural abnormalities while reducing computational costs. Experimental results on both the diabetic retinopathy (DR) dataset and ocular disease (OD) dataset demonstrate the superiority of the proposed method: it achieves a high accuracy of 0.9697 on the DR dataset and 0.9669 on the OD dataset, outperforming traditional methods such as CNN_eye, VGG, and UNet by more than 1 percentage point. Moreover, its training time is only half that of U-Net (on DR dataset) and VGG (on OD dataset), highlighting its potential for clinical DR screening. Full article

Penetration depth is a critical factor determining joint strength in butt welding; however, it is difficult to monitor in keyhole-mode laser welding due to the dynamic nature of the keyhole. Recently, optical coherence tomography (OCT) has been introduced for real-time keyhole depth measurement, though accurate results require meticulous calibration. In this study, deep learning-based models were developed to estimate penetration depth in laser welding of 780 dual-phase (DP) steel. The models utilized coaxial weld pool images and spectrometer signals as inputs, with OCT signals serving as the output reference. Both uni-sensor models (based on coaxial pool images) and multi-sensor models (incorporating spectrometer data) were developed using transfer learning techniques based on pre-trained convolutional neural network (CNN) architectures including MobileNetV2, ResNet50V2, EfficientNetB3, and Xception. The coefficients of determination values (R²) of the uni-sensor CNN transfer learning models without fine-tuning ranged from 0.502 to 0.681, and the mean absolute errors (MAEs) ranged from 0.152 mm to 0.196 mm. In the fine-tuning models, R² decreased by more than 17%, and MAE increased by more than 11% compared to the previous models without fine-tuning. In addition, in the multi-sensor model, R² ranged from 0.900 to 0.956, and MAE ranged from 0.058 mm to 0.086 mm, showing better performance than uni-sensor CNN transfer learning models. This study demonstrated the potential of using CNN transfer learning models for predicting penetration depth in laser welding of 780DP steel. Full article

Automatic skin lesion segmentation is essential for early melanoma diagnosis, yet the scarcity and limited diversity of annotated training data hinder progress. We introduce a two-stage framework that first employs a denoising diffusion probabilistic model (DDPM) enhanced with dilated convolutions and self-attention to synthesize unseen, high-fidelity dermoscopic images. In the second stage, segmentation models—including a dilated U-Net variant that leverages dilated convolutions to enlarge the receptive field—are trained on the augmented dataset. Experimental results demonstrate that this approach not only enhances segmentation accuracy across various architectures with an increase in DICE of more than 0.4, but also enables compact and computationally efficient segmentation models to achieve performance comparable to or even better than that of models with 10 times the parameters. Moreover, our diffusion-based data augmentation strategy consistently improves segmentation performance across multiple architectures, validating its effectiveness for developing accurate and deployable clinical tools. Full article

In maritime blurred image recognition, ship collision accidents frequently result from three primary blur types: (1) motion blur from vessel movement in complex sea conditions, (2) defocus blur due to water vapor refraction, and (3) scattering blur caused by sea fog interference. This paper proposes a dual-branch recognition method specifically designed for motion blur, which represents the most prevalent blur type in maritime scenarios. Conventional approaches exhibit constrained computational efficiency and limited adaptability across different modalities. To overcome these limitations, we propose a hybrid CNN–Transformer architecture: the CNN branch captures local blur characteristics, while the enhanced Transformer module models long-range dependencies via attention mechanisms. The CNN branch employs a lightweight ResNet variant, in which conventional residual blocks are substituted with Multi-Scale Gradient-Aware Residual Block (MSG-ARB). This architecture employs learnable gradient convolution for explicit local gradient feature extraction and utilizes gradient content gating to strengthen blur-sensitive region representation, significantly improving computational efficiency compared to conventional CNNs. The Transformer branch incorporates a Hierarchical Swin Transformer (HST) framework with Shifted Window-based Multi-head Self-Attention for global context modeling. The proposed method incorporates blur invariant Positional Encoding (PE) to enhance blur spectrum modeling capability, while employing DyT (Dynamic Tanh) module with learnable α parameters to replace traditional normalization layers. This architecture achieves a significant reduction in computational costs while preserving feature representation quality. Moreover, it efficiently computes long-range image dependencies using a compact 16 × 16 window configuration. The proposed feature fusion module synergistically integrates CNN-based local feature extraction with Transformer-enabled global representation learning, achieving comprehensive feature modeling across different scales. To evaluate the model’s performance and generalization ability, we conducted comprehensive experiments on four benchmark datasets: VAIS, GoPro, Mini-ImageNet, and Open Images V4. Experimental results show that our method achieves superior classification accuracy compared to state-of-the-art approaches, while simultaneously enhancing inference speed and reducing GPU memory consumption. Ablation studies confirm that the DyT module effectively suppresses outliers and improves computational efficiency, particularly when processing low-quality input data. Full article

Durian is a major tropical crop in Southeast Asia, but its yield and quality are severely impacted by a range of pests and diseases. Manual inspection remains the dominant detection method but suffers from high labor intensity, low accuracy, and difficulty in scaling. To address these challenges, this paper proposes EN-YOLO, a novel enhanced YOLO-based deep learning model that integrates the EfficientNet backbone and multimodal attention mechanisms for precise detection of durian pests and diseases. The model removes redundant feature layers and introduces a large-span residual edge to preserve key spatial information. Furthermore, a multimodal input strategy—incorporating RGB, near-infrared and thermal imaging—is used to enhance robustness under variable lighting and occlusion. Experimental results on real orchard datasets demonstrate that EN-YOLO outperforms YOLOv8 (You Only Look Once version 8), YOLOv5-EB (You Only Look Once version 5—Efficient Backbone), and Fieldsentinel-YOLO in detection accuracy, generalization, and small-object recognition. It achieves a 95.3% counting accuracy and shows superior performance in ablation and cross-scene tests. The proposed system also supports real-time drone deployment and integrates an expert knowledge base for intelligent decision support. This work provides an efficient, interpretable, and scalable solution for automated pest and disease management in smart agriculture. Full article

Prolonged operation of biomedical devices may compromise electronic component integrity due to cyclic thermal stress, thereby impacting both functionality and safety. Regulatory standards require regular inspections, particularly for surgical applications, highlighting the need for efficient and non-invasive diagnostic tools. This study introduces an integrated system that combines finite element models, infrared thermographic analysis, and artificial intelligence to monitor thermal stress in printed circuit boards (PCBs) within biomedical devices. A dynamic thermal model, implemented in COMSOL Multiphysics^® (version 6.2), identifies regions at high risk of thermal overload. The infrared measurements acquired through a FLIR P660 thermal camera provided experimental validation and a dataset for training a hybrid artificial intelligence system. This model integrates deep learning-based U-Net architecture for thermal anomaly segmentation with machine learning classification of heat diffusion patterns. By combining simulation, the proposed system achieved an F1-score of 0.970 for hotspot segmentation using a U-Net architecture and an F1-score of 0.933 for the classification of heat propagation modes via a Multi-Layer Perceptron. This study contributes to the development of intelligent diagnostic tools for biomedical electronics by integrating physics-based simulation and AI-driven thermographic analysis, supporting automatic classification and localisation of thermal anomalies, real-time fault detection and predictive maintenance strategies. Full article

To address the impact of distributed energy resource volatility on distribution network fault restoration, this paper proposes a strategy that incorporates net load forecasting. A Bayesian-optimized long short-term memory neural network is used to accurately predict the net load within fault-affected areas, achieving an R² of 0.9569 and an RMSE of 12.15 kW. Based on the forecasting results, a fast restoration optimization model is established, with objectives to maximize critical load recovery, minimize switching operations, and reduce network losses. The model is solved using a genetic algorithm enhanced with quantum particle swarm optimization (GA-QPSO), a hybrid metaheuristic known for its superior global exploration and local refinement capabilities. GA-QPSO has been successfully applied in various power system optimization problems, including service restoration, network reconfiguration, and distributed generation planning, owing to its effectiveness in navigating large, complex solution spaces. Simulation results on the IEEE 33-bus system show that the proposed method reduces network losses by 33.2%, extends the power supply duration from 60 to 120 min, and improves load recovery from 72.7% to 75.8%, demonstrating enhanced accuracy and efficiency of the restoration process. Full article

Peatlands are vital for global carbon cycling, and their ecological functions are influenced by vegetation composition. Accurate vegetation mapping is crucial for peatland management and conservation, but traditional methods face limitations such as low spatial resolution and labor-intensive fieldwork. We used ultra-high-resolution UAV imagery captured across seasonal and topographic gradients and assessed the impact of phenology and topography on classification accuracy. Additionally, this study evaluated the performance of four deep learning models (ResNet, Swin Transformer, ConvNeXt, and EfficientNet) for mapping vegetation in a subtropical Sphagnum peatland. ConvNeXt achieved peak accuracy at 87% during non-growing seasons through its large-kernel feature extraction capability, while ResNet served as the optimal efficient alternative for growing-season applications. Non-growing seasons facilitated superior identification of Sphagnum and monocotyledons, whereas growing seasons enhanced dicotyledon distinction through clearer morphological features. Overall accuracy in low-lying humid areas was 12–15% lower than in elevated terrain due to severe spectral confusion among vegetation. SHapley Additive exPlanations (SHAP) of the ConvNeXt model identified key vegetation indices, the digital surface model, and select textural features as primary performance drivers. This study concludes that the combination of deep learning and UAV imagery presents a powerful tool for peatland vegetation mapping, highlighting the importance of considering phenological and topographical factors. Full article