Search Results (243)

During the cold metal transfer plus pulse (CMT+P) welding process of medium-thick plates, problems such as incomplete penetration (IP) and burn-through (BT) are prone to occur, and weld pool morphology is important information reflecting the penetration states. In order to acquire high-quality weld pool images under complex welding conditions, such as smoke and arc light, a welding monitoring system was designed. For the purpose of predicting weld penetration states, the improved Inception-ResNet prediction model was proposed. Squeeze-and-Excitation (SE) block was added after each Inception-ResNet block to further extract key feature information from weld pool images, increasing the weight of key features beneficial for predicting the penetration states. The model has been trained, validated, and tested. The results demonstrate that the improved model has an accuracy of over 96% in predicting penetration states of aluminum alloy medium-thick plates compared to the original model. The model was applied in welding experiments and achieved an accurate prediction. Full article

With advancements in IoT and smart home tech, WiFi-driven gesture recognition is attracting more focus due to its non-contact nature and user-friendly design. However, WiFi signals are affected by multipath effects, attenuation, and interference, resulting in complex and variable signal patterns that pose challenges for accurately modeling gesture characteristics. This study proposes SE-WiGR, an innovative WiFi gesture recognition method to address these challenges. First, channel state information (CSI) related to gesture actions is collected using commercial WiFi devices. Next, the data is preprocessed, and Doppler-shift image data is extracted as input for the network model. Finally, the method integrates the squeeze-and-excitation (SE) mechanism with the VGG16 network to classify gestures. The method achieves a recognition accuracy of 94.12% across multiple scenarios, outperforming the standalone VGG16 network by 4.13%. This improvement confirms that the SE module effectively enhances gesture feature extraction while suppressing background noise. Full article

Aiming at the problem that the existing auxiliary diagnosis methods for fractures are mostly limited to specific body parts and lack generality and robustness when applied to multi-part diagnoses, this study proposes a two-stage upper limb fracture auxiliary diagnosis method based on deep learning and develops a corresponding auxiliary diagnosis system. In the first stage, this study employs an improved ResNet-50 model combined with transfer learning and a Squeeze-and-Excitation (SE) attention mechanism for fracture image localization. In the second stage, an improved You Only Look Once (YOLO) model based on Scale Sequence Feature Fusion (SSFF) and Triple Feature Encoder (TFE) modules is used for fracture diagnoses in different body parts. Contrary to the traditional methods that are tailored to specific body parts, the integrated design approach presented in this paper is better suited to meeting the diagnostic needs of multiple body parts, demonstrating better generality and clinical application potential. Full article

Phyllosticta fragaricola-induced angular leaf spot causes substantial economic losses in global strawberry production, necessitating advanced severity assessment methods. This study proposed a dual-phase grading framework integrating deep learning and computer vision. The enhanced You Only Look Once version 11 (YOLOv11) architecture incorporated a Content-Aware ReAssembly of FEatures (CARAFE) module for improved feature upsampling and a squeeze-and-excitation (SE) attention mechanism for channel-wise feature recalibration, resulting in the YOLOv11-CARAFE-SE for the severity assessment of strawberry angular leaf spot. Furthermore, an OpenCV-based threshold segmentation algorithm based on H-channel thresholds in the HSV color space achieved accurate lesion segmentation. A disease severity grading standard for strawberry angular leaf spot was established based on the ratio of lesion area to leaf area. In addition, specialized software for the assessment of disease severity was developed based on the improved YOLOv11-CARAFE-SE model and OpenCV-based algorithms. Experimental results show that compared with the baseline YOLOv11, the performance is significantly improved: the box mAP@0.5 is increased by 1.4% to 93.2%, the mask mAP@0.5 is increased by 0.9% to 93.0%, the inference time is shortened by 0.4 ms to 0.9 ms, and the computational load is reduced by 1.94% to 10.1 GFLOPS. In addition, this two-stage grading framework achieves an average accuracy of 94.2% in detecting selected strawberry horn leaf spot disease samples, providing real-time field diagnostics and a high-throughput phenotypic analysis for resistance breeding programs. This work demonstrates the feasibility of rapidly estimating the severity of strawberry horn leaf spot, which will establish a robust technical framework for strawberry disease management under field conditions. Full article

The integration of Context-Aware Systems (CASs) in Future Operating Rooms (FORs) aims to enhance surgical workflows and outcomes through real-time data analysis. CASs require accurate classification of surgical tools, enabling the understanding of surgical actions. This study proposes a novel deep learning approach for surgical tool classification based on combining convolutional neural networks (CNNs), Feature Fusion Modules (FFMs), Squeeze-and-Excitation (SE) networks, and Bidirectional long-short term memory (BiLSTM) networks to capture both spatial and temporal features in laparoscopic surgical videos. We explored different modeling scenarios with respect to the location and number of SE blocks for multi-label surgical tool classification in the Cholec80 dataset. Furthermore, we analyzed a single-label surgical tool classification model using a simplified and computationally less expensive architecture compared to the multi-label problem setting. The single-label classification model showed an improved overall performance compared to the proposed multi-label classification model due to the increased complexity of identifying multiple tools simultaneously. Nonetheless, our results demonstrated that the proposed CNN-SE-FFM-BiLSTM multi-label model achieved competitive performance to state-of-the-art methods with excellent performance in detecting tools with complex usage patterns and in minority classes. Future work should focus on optimizing models for real-time applications, and broadening dataset evaluations to improve performance in diverse surgical environments. These improvements are crucial for the practical implementation of such models in CASs, ultimately aiming to enhance surgical workflows and patient outcomes in FORs. Full article

Seafloor topography super-resolution reconstruction is critical for marine resource exploration, geological monitoring, and navigation safety. However, sparse acoustic data frequently result in the loss of high-frequency details, and traditional deep learning models exhibit limitations in uncertainty quantification, impeding their practical application. To address these challenges, this study systematically investigates the combined effects of various regularization strategies and uncertainty quantification modules. It proposes a hybrid dropout model that jointly optimizes high-precision reconstruction and uncertainty estimation. The model integrates residual blocks, squeeze-and-excitation (SE) modules, and a multi-scale feature extraction network while employing Monte Carlo Dropout (MC-Dropout) alongside heteroscedastic noise modeling to dynamically gate the uncertainty quantification process. By adaptively modulating the regularization strength based on feature activations, the model preserves high-frequency information and accurately estimates predictive uncertainty. The experimental results demonstrate significant improvements in the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Peak Signal-to-Noise Ratio (PSNR). Compared to conventional dropout architectures, the proposed method achieves a PSNR increase of 46.5% to 60.5% in test regions with a marked reduction in artifacts. Overall, the synergistic effect of employed regularization strategies and uncertainty quantification modules substantially enhances detail recovery and robustness in complex seafloor topography reconstruction, offering valuable theoretical insights and practical guidance for further optimization of deep learning models in challenging applications. Full article

The detection of equipment defects in substations is crucial for maintaining the normal operation of power systems. This paper proposes an object detection algorithm for substation equipment defect detection based on improvements to the YOLOv8 model. First, the backbone of YOLOv8 is replaced with EfficientViT, which not only reduces computational redundancy but also enhances the model’s feature extraction capabilities, thereby improving overall performance. Second, a Squeeze-and-Excitation (SE) attention mechanism module is incorporated at the terminal stage of the backbone network to reinforce channel-wise feature representation in input feature maps. Finally, the Bottleneck component within YOLOv8’s C2f module is substituted with FasterBlock, which significantly accelerates inference speed while maintaining model accuracy. Experimental results on the substation equipment defect dataset demonstrate that the improved algorithm achieves a mean average precision (mAP) of 92.8%, representing a 1.8% enhancement over the baseline model. The substantial improvement in average precision confirms the feasibility and effectiveness of the proposed modifications to the YOLOv8 architecture. Full article

During rice cultivation, extracting levees helps to delineate effective planting areas, thereby enhancing the precision of management zones. This approach is crucial for devising more efficient water field management strategies and has significant implications for water-saving irrigation and fertilizer optimization in rice production. The uneven distribution and lack of standardization of levees pose significant challenges for their accurate extraction. However, recent advancements in remote sensing and deep learning technologies have provided viable solutions. In this study, Youyi Farm in Shuangyashan City, Heilongjiang Province, was chosen as the experimental site. We developed the SCA-UNet model by optimizing the UNet algorithm and enhancing its network architecture through the integration of the Convolutional Block Attention Module (CBAM) and Squeeze-and-Excitation Networks (SE). The SCA-UNet model leverages the channel attention strengths of SE while incorporating CBAM to emphasize spatial information. Through a dual-attention collaborative mechanism, the model achieves a synergistic perception of the linear features and boundary information of levees, thereby significantly improving the accuracy of levee extraction. The experimental results demonstrate that the proposed SCA-UNet model and its additional modules offer substantial performance advantages. Our algorithm outperforms existing methods in both computational efficiency and precision. Significance analysis revealed that our method achieved overall accuracy (OA) and F1-score values of 88.4% and 90.6%, respectively. These results validate the efficacy of the multimodal dataset in addressing the issue of ambiguous levee boundaries. Additionally, ablation experiments using 10-fold cross-validation confirmed the effectiveness of the proposed SCA-UNet method. This approach provides a robust technical solution for levee extraction and has the potential to significantly advance precision agriculture. Full article

In actual operation, the output power of distributed marine photovoltaic monitoring faces challenges from wind, waves, and other dynamic motion factors. To address these challenges, this paper proposes a novel maximum power point inference method for distributed marine photovoltaic monitoring. First, a digital fusion model has been constructed to obtain a comprehensive dataset of the distributed marine photovoltaic monitoring system. Second, Multilayer Convolutional Neural Networks (CNN) are constructed to extract the local high-frequency motion characteristics, Squeeze and Excitation Attention (SE-Attention) is employed to capture the global low-frequency motion characteristics, and Long Short-Term Memory (LSTM) is utilized to perform temporal modeling of the motion characteristics. Subsequently, the Crested Porcupine Optimizer (CPO) algorithm is used to achieve high-precision recognition of the maximum power point in distributed marine photovoltaic monitoring. Finally, the effectiveness of the method is verified through experiments and simulations. The results indicate that the maximum power point of distributed marine photovoltaic monitoring exhibits multi-spectral motion characteristics, with the highest frequency at 335.2 Hz and the lowest frequency at 12.9 Hz. The proposed method enables efficient inference of the maximum power point for distributed marine photovoltaic monitoring under motion conditions, with an accuracy of 98.63%. Full article

Human Pose Estimation (HPE) remains challenging due to scale variation, occlusion, and high computational costs. Standard methods often struggle to capture detailed spatial information when keypoints are obscured, and they typically rely on computationally expensive deconvolution layers for upsampling, making them inefficient for real-time or resource-constrained scenarios. We propose AMFACPose (Attentive Multi-scale Features with Adaptive Context PoseResNet) to address these limitations. Specifically, our architecture incorporates Coordinate Convolution 2D (CoordConv2d) to retain explicit spatial context, alleviating the loss of coordinate information in conventional convolutions. To reduce computational overhead while maintaining accuracy, we utilize Depthwise Separable Convolutions (DSCs), separating spatial and pointwise operations. At the core of our approach is an Adaptive Feature Pyramid Network (AFPN), which replaces costly deconvolution-based upsampling by efficiently aggregating multi-scale features to handle diverse human poses and body sizes. We further introduce Dual-Gate Context Blocks (DGCBs) that refine global context to manage partial occlusions and cluttered backgrounds. The model integrates Squeeze-and-Excitation (SE) blocks and the Spatial–Channel Refinement Module (SCRM) to emphasize the most informative feature channels and spatial regions, which is particularly beneficial for occluded or overlapping keypoints. For precise keypoint localization, we replace dense heatmap predictions with coordinate classification using Multi-Layer Perceptron (MLP) heads. Experiments on the COCO and CrowdPose datasets demonstrate that AMFACPose surpasses the existing 2D HPE methods in both accuracy and computational efficiency. Moreover, our implementation on edge devices achieves real-time performance while preserving high accuracy, confirming the suitability of AMFACPose for resource-constrained pose estimation in both benchmark and real-world environments. Full article

This study proposes an intelligent agricultural pest monitoring system that integrates mechanical control with deep learning to address issues in traditional systems, such as pest accumulation interference, image contrast degradation under complex lighting, and poor balance between model accuracy and real-time performance. A three-axis coordinated separation device is employed, achieving a 92.41% single-attempt separation rate and 98.12% after three retries. Image preprocessing combines the Multi-Scale Retinex with Color Preservation (MSRCP) algorithm and bilateral filtering to enhance illumination correction and reduce noise. For overlapping pest detection, EfficientNetv2-S replaces the YOLOv5s backbone and is combined with an Adaptive Feature Pyramid Network (AFPN), achieving 95.72% detection accuracy, 94.04% mAP, and 127 FPS. For pest species recognition, the model incorporates a Squeeze-and-Excitation (SE) attention module and α-CIoU loss function, reaching 91.30% precision on 3428 field images. Deployed on an NVIDIA Jetson Nano, the system demonstrates a detection time of 0.3 s, 89.64% recall, 86.78% precision, and 1.136 s image transmission delay, offering a reliable solution for real-time pest monitoring in complex field environments. Full article

Cracks and dents in concrete structures are core defects that threaten building safety, but the existing YOLO series algorithms face a huge bottleneck in complex engineering scenarios. Tiny cracks are susceptible to background texture interference, leading to misjudgment. The traditional detection frame has difficulty in accurately characterizing the dent geometry, which affects the quantitative damage assessment. In this paper, we propose a Multi-level Defect Fusion Segmentation Network (MDFNet) to break through the single-task limitation through the detection segmentation synergy framework. We improve the anchor frame strategy of YOLOv11 and enhance the recall of small targets by combining Copy–Pasting, and then enhance the pixel-level characterization of crack edges and dent contours by embedding the Head Attention-Expanded Convolutional Fusion Module (HAEConv) in U-Net with squeeze-and-excitation (SE) channel attention. Joint detection loss and segmentation loss are used for task co-optimization. On our self-constructed concrete defect dataset, MDFNet significantly outperforms the baseline model. In terms of accuracy, the MDFNet Dice coefficient is 92.4%, an improvement of 4.1 percentage points compared to YOLOv11-Seg. Our mean Intersection over Union (mIoU) reaches 81.6%, with strong generalization ability under complex background interference. In terms of engineering efficacy, the model achieves a processing speed of 45 frames per second (FPS) for 640 × 640 images, which is able to meet real-time monitoring requirements. The experimental results verify the feasibility of the model in the research field of crack and dent detection in concrete structures. Full article

Temperature plays a crucial role in the grain storage process and food security. Due to limitations in grain storage data acquisition in real-world scenarios, this paper proposes a data augmentation method for grain storage data that operates in both the time and frequency domains, as well as an enhanced grain storage temperature prediction model. To address the issue of small sample sizes in grain storage temperature data, Gaussian noise is added to the grain storage temperature data in the time domain to highlight the subtle variations in the original data. The fast Fourier transform (FFT) is employed in the frequency domain to highlight periodicity and trends in the grain storage temperature data. The prediction model uses a long short-term memory (LSTM) network, enhanced with convolution layers for feature extraction and a Squeeze-and-Excitation Networks (SENet) module to suppress unimportant features and highlight important ones. Experimental results show that the FTA-CNN-SE-LSTM compares with the original LSTM network, and the MAE and RMSE are reduced by 74.77% and 74.02%, respectively. It solves the problem of data limitation in the actual grain storage process, greatly improves the accuracy of grain storage temperature prediction, and can accurately prevent problems caused by abnormal grain pile temperature. Full article

To address the nonlinear and non-stationary characteristics of dissolved gas concentration data in transformer oil, this paper proposes a hybrid prediction model (VMD-SSA-LSTM-SE) that integrates Variational Mode Decomposition (VMD), the Whale Optimization Algorithm (WOA), the Sparrow Search Algorithm (SSA), Long Short-Term Memory (LSTM), and the Squeeze-and-Excitation (SE) attention mechanism. First, WOA dynamically optimizes VMD parameters (mode number k and penalty factor α to effectively separate noise and valid signals, avoiding modal aliasing). Then, SSA globally searches for optimal LSTM hyperparameters (hidden layer nodes, learning rate, etc.) to enhance feature mining for non-continuous data. The SE attention mechanism recalibrates channel-wise feature weights to capture critical time-series patterns. Experimental validation using real transformer oil data demonstrates that the model outperforms existing methods in prediction accuracy and computational efficiency. For instance, the CH₄ test set achieves a Mean Absolute Error (MAE) of 0.17996 μL/L, a Mean Absolute Percentage Error (MAPE) of 1.4423%, and an average runtime of 82.7 s, making it significantly faster than CEEMDAN-based models. These results provide robust technical support for transformer fault prediction and condition-based maintenance, highlighting the model’s effectiveness in handling non-stationary time-series data. Full article

Driver distraction has become a critical factor in traffic accidents, necessitating accurate behavior recognition for road safety. However, existing methods still suffer from limitations such as low accuracy in recognizing drivers’ localized actions and difficulties in distinguishing subtle differences between different behaviors. This paper proposes RMTSE, a hybrid attention model, to enhance driver distraction recognition. The framework introduces a Manhattan Self-Attention Squeeze-and-Excitation (MaSA-SE) module that combines spatial self-attention with channel attention mechanisms. This integration enables simultaneous enhancement of discriminative features and suppression of irrelevant characteristics in driving behavior images, improving learning efficiency through focused feature extraction. We also propose to employ a transfer learning strategy utilizing pre-trained weights during the training process, which further accelerates model convergence and enhances feature generalization. The model achieves Top-1 accuracies of 99.82% and 94.95% on SFD3 and 100-Driver datasets, respectively, with minimal parameter increments, outperforming existing state-of-the-art methods. Full article