Search Results (60)

To address challenges such as small target size, high density, severe occlusion, complex background interference, and edge device computational constraints, a lightweight model, GAFNet, is proposed based on YOLO11n, optimized for rice pest detection in field environments. To improve feature perception, we propose the Global Attention Fusion and Spatial Pyramid Pooling (GAM-SPP) module, which captures global context and aggregates multi-scale features. Building on this, we introduce the C3-Efficient Feature Selection Attention (C3-EFSA) module, which refines feature representation by combining depthwise separable convolutions (DWConv) with lightweight channel attention to enhance background discrimination. The model’s detection head, Enhanced Ghost Detect (EGDetect), integrates Enhanced Ghost Convolution (EGConv), Squeeze-and-Excitation (SE), and Sigmoid-Weighted Linear Unit (SiLU) activation, which reduces redundancy. Additionally, we propose the Focal-Enhanced Complete-IoU (FECIoU) loss function, incorporating stability and hard-sample weighting for improved localization. Compared to YOLO11n, GAFNet improves Precision, Recall, and mean Average Precision (mAP) by 3.5%, 4.2%, and 1.6%, respectively, while reducing parameters and computation by 5% and 21%. GAFNet can deploy on edge devices, providing farmers with instant pest alerts. Further, GAFNet is evaluated on the AgroPest-12 dataset, demonstrating enhanced generalization and robustness across diverse pest detection scenarios. Overall, GAFNet provides an efficient, reliable, and sustainable solution for early pest detection, precision pesticide application, and eco-friendly pest control, advancing the future of smart agriculture. Full article

Additive manufacturing (AM) has emerged as a pivotal technology in component fabrication, renowned for its capabilities in freeform fabrication, material efficiency, and integrated design-to-manufacturing processes. As a critical branch of AM, metal additive manufacturing (MAM) has garnered significant attention for producing metal parts. However, process anomalies during MAM can pose safety risks, while internal defects in as-built parts detrimentally affect their service performance. These concerns underscore the necessity for robust in-process monitoring of both the MAM process and the quality of the resulting components. This review first delineates common MAM techniques and popular in-process monitoring methods. It then elaborates on the fundamental principles of acoustic emission (AE), including the configuration of AE systems and methods for extracting characteristic AE parameters. The core of the review synthesizes applications of AE technology in MAM, categorizing them into three key aspects: (1) hardware setup, which involves a comparative analysis of sensor selection, mounting strategies, and noise suppression techniques; (2) parametric characterization, which establishes correlations between AE features and process dynamics (e.g., process parameter deviations, spattering, melting/pool stability) as well as defect formation (e.g., porosity and cracking); and (3) intelligent monitoring, which focuses on the development of classification models and the integration of feedback control systems. By providing a systematic overview, this review aims to highlight the potential of AE as a powerful tool for real-time quality assurance in MAM. Full article

Net ecosystem exchange (NEE) is a central metric for assessing carbon cycling, and its accurate quantification is critical for understanding terrestrial-atmosphere carbon exchange dynamics. However, in complex alpine regions, high-resolution NEE estimation remains challenging due to limited observations and heterogeneous surface processes. To address this, we developed a multimodal feature fusion model (Multimodal-CNN-Attention-RF, MMCA-RF) that integrates convolutional neural networks (CNN) and random forest (RF) for NEE estimation in the Babao River Basin on the northeastern Tibetan Plateau. The model incorporates a cross-modal attention mechanism to dynamically optimize feature interactions, thereby better capturing the spatially heterogeneous responses of vegetation to environmental drivers. Results demonstrate that MMCA-RF exhibits strong stability and generalization, with R² values of 0.89 (training) and 0.85 (testing). Based on model outputs, the Babao River Basin acted as a carbon sink during 2017–2023, with a mean annual NEE of −100.86 gC m⁻² yr⁻¹. Spatially, NEE showed pronounced heterogeneity, while seasonal variation followed a unimodal pattern. Among vegetation types, grasslands contributed the largest total carbon sink, whereas open woodlands showed the highest sequestration efficiency per unit area. Driver analysis identified temperature as the dominant control on NEE spatial variation, with interactions between temperature, precipitation, and topography further enhancing heterogeneity. This study provides a high-accuracy modeling approach for monitoring carbon cycling in alpine ecosystems and offers insights into the stability of regional carbon pools under climate change. Full article

Predictive maintenance for marine engines requires forecasts that are both accurate and technically interpretable. This work introduces BEACON, a hybrid architecture that combines a bidirectional long short-term memory encoder with attention pooling, a Kolmogorov Arnold network and a lightweight multilayer perceptron for cylinder-level exhaust gas temperature forecasting, evaluated in both centralized and federated learning settings. On operational data from a bulk carrier, BEACON outperformed strong state-of-the-art baselines, achieving an RMSE of $0.5905$, MAE of $0.4713$ and $R^2$ of approximately 0.95, while producing interpretable response curves and stable SHAP rankings across engine load regimes. A second contribution is the explicit evaluation of explanation stability in a federated learning setting, where BEACON maintained competitive accuracy and attained mean Spearman correlations above $0.8$ between client-specific SHAP rankings, whereas baseline models exhibited substantially lower agreement. These results indicate that the proposed hybrid design provides an accurate and explanation-stable foundation for privacy-aware predictive maintenance of marine engines. Full article

Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention (DMA) mechanism, enabling effective cross-modal knowledge transfer for capturing macro- and micro-anomalies. A Stabilized Attention-based Pooling (SAP) module adaptively aggregates discriminative representations using self-generated anomaly heatmaps, enhancing localization accuracy and mitigating feature dilution. Trained solely in auxiliary datasets with multi-task segmentation and contrastive losses, the approach requires no target-domain samples. Extensive evaluation across seven benchmarks (MVTec AD, VisA, BTAD, MPDD, KSDD, DAGM, DTD-Synthetic) demonstrates state-of-the-art performance, achieving 93.4% image-level AUROC, 94.3% AP, 96.9% pixel-level AUROC, and 92.4% AUPRO on average. Ablation studies confirm the efficacy of DMA and SAP, while qualitative results highlight superior boundary precision and noise suppression. The framework offers a scalable, annotation-efficient solution for real-world industrial anomaly detection. Full article

Speech Emotion Recognition (SER) is important for applications such as Human–Computer Interaction (HCI) and emotion-aware services. Traditional SER models rely on utterance-level labels, aggregating frame-level representations through pooling operations. However, emotional states can vary across frames within an utterance, making it difficult for models to learn consistent and robust representations. To address this issue, we propose two auxiliary loss functions, Emotional Attention Loss (EAL) and Frame-to-Utterance Alignment Loss (FUAL). The proposed approach uses a Classification token (CLS) self-attention pooling mechanism, where the CLS summarizes the entire utterance sequence. EAL encourages frames of the same emotion to align closely with the CLS while separating frames of different classes, and FUAL enforces consistency between frame-level and utterance-level predictions to stabilize training. Model training proceeds in two stages: Stage 1 fine-tunes the wav2vec 2.0 backbone with Cross-Entropy (CE) loss to obtain stable frame embeddings, and stage 2 jointly optimizes CE, EAL and FUAL within the CLS-based pooling framework. Experiments on the IEMOCAP four-class dataset demonstrate that our method consistently outperforms baseline models, showing that the proposed losses effectively address representation inconsistencies and improve SER performance. This work advances Artificial Intelligence by improving the ability of models to understand human emotions through speech. Full article

Stomach ulcers, a common type of gastrointestinal (GI) disease, pose serious health risks if not diagnosed and treated at an early stage. Therefore, in this research, a method is proposed based on two deep learning models for classification and segmentation. The classification model is based on Convolutional Neural Networks (CNN) and incorporates Q-learning to achieve learning stability and decision accuracy through reinforcement-based feedback. In this model, input images are passed through a custom CNN model comprising seven layers, including convolutional, ReLU, max pooling, flattening, and fully connected layers, for feature extraction. Furthermore, the agent selects an action (class) for each input and receives a +1 reward for a correct prediction and −1 for an incorrect one. The Q-table stores a mapping between image features (states) and class predictions (actions), and is updated at each step based on the reward using the Q-learning update rule. This process runs over 1000 episodes and utilizes Q-learning parameters (α = 0.1, γ = 0.6, ϵ = 0.1) to help the agent learn an optimal classification strategy. After training, the agent is evaluated on the test data using only its learned policy. The classified ulcer images are passed to the proposed attention-based U-Net model to segment the lesion regions. The model contains an encoder, a decoder, and attention layers. The encoder block extracts features through pooling and convolution layers, while the decoder block up-samples the features and reconstructs the segmentation map. Similarly, the attention block is used to highlight the important features obtained from the encoder block before passing them to the decoder block, helping the model focus on relevant spatial information. The model is trained using the selected hyperparameters, including an 8-batch size, the Adam optimizer, and 50 epochs. The performance of the models is evaluated on Kvasir, Nerthus, CVC-ClinicDB, and a private POF dataset. The classification framework provides 99.08% accuracy on Kvasir and 100% accuracy on Nerthus. In contrast, the segmentation framework yields 98.09% accuracy on Kvasir, 99.77% accuracy on Nerthus, 98.49% accuracy on CVC-ClinicDB, and 99.13% accuracy on the private dataset. The achieved results are superior to those of previous methods published in this domain. Full article

Accurate angular alignment of wafers is essential in ion implantation to prevent channeling effects that degrade device performance. This study proposes a real-time notch-angle-detection system based on you only look once version 8 with oriented bounding boxes (YOLOv8-OBB). The proposed method compares YOLOv8 and YOLOv8-OBB, demonstrating the superiority of the latter in accurately capturing rotational features. To enhance detection performance, hyperparameters—including initial learning rate (Lr0), weight decay, and optimizer—are optimized using an one factor at a time (OFAT) approach followed by grid search. Architectural improvements, including spatial pyramid pooling fast with large selective kernel attention (SPPF_LSKA), a bidirectional feature pyramid network (BiFPN), and a high-resolution detection head (P2 head), are incorporated to improve small-object detection. Furthermore, a gradual unfreezing strategy is employed to support more effective and stable transfer learning. The final system is evaluated over 100 training epochs and tracked up to 5000 epochs to verify long-term stability. Compared to baseline models, it achieves higher accuracy and robustness in angle-sensitive scenarios, offering a reliable and scalable solution for high-precision wafer-notch detection in semiconductor manufacturing. Full article

Early pest detection in edible crops demands sensing solutions that can run at the edge under tight power, budget, and maintenance constraints. This review synthesizes peer-reviewed work (2015–2025) on three modality families—vision/AI, spectroscopy/imaging spectroscopy, and indirect sensors—restricted to edible crops and studies reporting some implementation or testing (n = 178; IEEE Xplore and Scopus). Each article was scored with a modality-aware performance–cost–implementability (PCI) rubric using category-specific weights, and the inter-reviewer reliability was quantified with weighted Cohen’s $\kappa$. We translated the evidence into compact decision maps for common deployment profiles (low-power rapid rollout; high-accuracy cost-flexible; and block-scale scouting). Across the corpus, vision/AI and well-engineered sensor systems more often reached deployment-leaning PCI (≥3.5: 32.0% and 33.3%, respectively) than spectroscopy (18.2%); the median PCI was 3.20 (AI), 3.17 (sensors), and 2.60 (spectroscopy). A Pareto analysis highlighted detector/attention models near $(P,C,I)\approx (4,5,4)$; sensor nodes spanning balanced $(4,4,4)$ and ultra-lean $(2,5,4)$ trade-offs; and the spectroscopy split between the early-warning strength $(5,4,3)$ and portability $(4,3,4)$. The inter-rater agreement was substantial for sensors and spectroscopy (pooled quadratic $\kappa$ = 0.73–0.83; up to 0.93 by dimension) and modest for imaging/AI (PA vs. Author 2: ${\kappa }_{\mathrm{quadratic}}=0.30$–$0.44$), supporting rubric stability with adjacency-dominated disagreements. The decision maps operationalize these findings, helping practitioners select a fit-for-purpose modality and encouraging a minimum PCI metadata set to enable reproducible, deployment-oriented comparisons. Full article

Flange bolts are vital fasteners in civil, mechanical, and aerospace structures, where preload stability directly affects overall safety. Conventional methods for bolt loosening detection often suffer from missed detections, weak feature representation, and insufficient cross-scale fusion under complex backgrounds. This paper presents an integrated detection and angle estimation framework using a lightweight deep learning detection network. A MobileViT backbone is employed to balance local texture with global context. In the spatial pyramid pooling stage, large separable convolutional kernels are combined with a channel and spatial attention mechanism to highlight discriminative features while suppressing noise. Together with content-aware upsampling and bidirectional multi-scale feature fusion, the network achieves high accuracy in detecting small and low-contrast targets while maintaining real-time performance. For angle estimation, the framework adopts an efficient training-free pipeline consisting of oriented FAST and rotated BRIEF feature detection, approximate nearest neighbor matching, and robust sample consensus fitting. This approach reliably removes false correspondences and extracts stable rotation components, maintaining success rates between 85% and 93% with an average error close to one degree, even under reflection, blur, or moderate viewpoint changes. Experimental validation demonstrates strong stability in detection and angular estimation under varying illumination and texture conditions, with a favorable balance between computational efficiency and practical applicability. This study provides a practical, intelligent, and deployable solution for bolt loosening detection, supporting the safe operation of large-scale equipment and infrastructure. Full article

Background/Objectives: Lightweight CNNs for medical imaging remain limited. We propose TurkerNeXtV2, a compact CNN that introduces two new blocks: a pooling-based attention with an inverted bottleneck (TNV2) and a hybrid downsampling module. These blocks improve stability and efficiency. The aim is to achieve transformer-level effectiveness while keeping the simplicity, low computational cost, and deployability of CNNs. Methods: The model was first pretrained on the Stable ImageNet-1k benchmark and then fine-tuned on a collected plantar-pressure OA dataset. We also evaluated the model on a public blood-cell image dataset. Performance was measured by accuracy, precision, recall, and F1-score. Inference time (images per second) was recorded on an RTX 5080 GPU. Grad-CAM was used for qualitative explainability. Results: During pretraining on Stable ImageNet-1k, the model reached a validation accuracy of 87.77%. On the OA test set, the model achieved 93.40% accuracy (95% CI: 91.3–95.2%) with balanced precision and recall above 90%. On the blood-cell dataset, the test accuracy was 98.52%. The average inference time was 0.0078 s per image (≈128.8 images/s), which is comparable to strong CNN baselines and faster than the transformer baselines tested under the same settings. Conclusions: TurkerNeXtV2 delivers high accuracy with low computational cost. The pooling-based attention (TNV2) and the hybrid downsampling enable a lightweight yet effective design. The model is suitable for real-time and clinical use. Future work will include multi-center validation and broader tests across imaging modalities. Full article

In recent years, fractional-order controllers have garnered increasing attention due to their enhanced flexibility and superior dynamic performance in control system design. Among them, the fractional-order Proportional–Integral–Derivative (FOPID) controller offers two additional tunable parameters, $\lambda$ and $\mu$, expanding the design space and allowing for finer performance tuning. However, the increased parameter dimensionality poses significant challenges for optimisation. To address this, the present study investigates the application of FOPID controllers to a two-wheeled self-balancing robot (TWSBR), with controller parameters optimised using intelligent algorithms. A novel Multi-Strategy Improved Beluga Whale Optimization (MSBWO) algorithm is proposed, integrating chaotic mapping, elite pooling, adaptive Lévy flight, and a golden sine search mechanism to enhance global convergence and local search capability. Comparative experiments are conducted using several widely known algorithms to evaluate performance. Results demonstrate that the FOPID controller optimised via the proposed MSBWO algorithm significantly outperforms both traditional PID and FOPID controllers tuned by other optimisation strategies, achieving faster convergence, reduced overshoot, and improved stability. Full article

Urban drainage pipelines are essential components of smart city infrastructure, supporting the safe and sustainable operation of underground systems. However, internal corrosion in pipelines poses significant risks to structural stability and public safety. In this study, we propose an enhanced semantic segmentation framework based on High-Resolution Network Version 2 (HRNetV2) to accurately identify corroded regions in Traditional closed-circuit television (CCTV) images. The proposed method integrates a Convolutional Block Attention Module (CBAM) to strengthen the feature representation of corrosion patterns and introduces a Lightweight Pyramid Pooling Module (LitePPM) to improve multi-scale context modeling. By preserving high-resolution details through HRNetV2’s parallel architecture, the model achieves precise and robust segmentation performance. Experiments on a real-world corrosion dataset show that our approach attains a mean Intersection over Union ($mIoU$) of 95.92 $\pm$ 0.03%, $Recall$ of 97.01 $\pm$ 0.02%, and an overall $Accuracy$ of 98.54%. These results demonstrate the method’s effectiveness in supporting intelligent infrastructure inspection and provide technical insights for advancing automated maintenance systems in smart cities. Full article

Aluminum foil sealing is widely employed in industrial packaging, and the quality of sealing plays a crucial role in ensuring product integrity and safety. Thermal infrared images frequently exhibit non-uniform heat distribution and indistinct boundaries within the sealing region. Additionally, variations in thermal response and local structural characteristics are observed across different defect types. Thus, traditional detection methods exhibit limitations regarding their stability and adaptability. In this paper, a novel thermal image recognition algorithm called EAC-Net is proposed for the classification and detection of sealing defects in thermal infrared images. In the proposed method, EfficientNet-B0 is utilized as the backbone network to improve its adaptability for industrial deployment. Furthermore, the Atrous Spatial Pyramid Pooling module is incorporated to enhance the multi-scale perception of defect regions, while the Channel–Spatial Attention Mixing with Channel Shuffle module is adopted to strengthen the focus on critical thermal features. Significant improvements in recognition performance were verified in experiments, while both computational complexity and inference latency were effectively kept at low levels. In the experiments, EAC-Net demonstrated an accuracy of 99.06% and a precision of 99.07%, indicating its high robustness and application potential. Full article

To address the limitations of CNNs and RNNs in handling complex operating conditions, multi-scale degradation patterns, and long-term dependencies—with attention mechanisms often failing to highlight key degradation features—this paper proposes a remaining useful life (RUL) prediction framework based on a multi-scale dilated fusion attention (MDFA) module. The MDFA leverages parallel dilated convolutions with varying dilation rates to expand receptive fields, while a global-pooling branch captures sequence-level degradation trends. Additionally, integrated channel and spatial attention mechanisms enhance the model’s ability to emphasize informative features and suppress noise, thereby improving overall prediction robustness. The proposed method is evaluated on NASA’s C-MAPSS and N-CMAPSS datasets, achieving MAE values of 0.018–0.026, RMSE values of 0.021–0.032, and R² scores above 0.987, demonstrating superior accuracy and stability compared to existing baselines. Furthermore, to verify generalization across domains, experiments on the PHM2012 bearing dataset show similar performance (MAE: 0.023–0.026, RMSE: 0.031–0.032, R²: 0.987–0.995), confirming the model’s effectiveness under diverse operating conditions and its adaptability to different degradation behaviors. This study provides a practical and interpretable deep-learning solution for RUL prediction, with broad applicability to aero-engine prognostics and other industrial health-monitoring tasks. Full article