Lightweight Model Research Articles

OMC-YOLO: A Lightweight Grading Detection Method for Oyster Mushrooms

In this paper, a lightweight model—OMC-YOLO, improved based on YOLOv8n—is proposed for the automated detection and grading of oyster mushrooms. Aiming at the problems of low efficiency, high costs, and the difficult quality assurance of manual operations in traditional oyster mushroom cultivation, OMC-YOLO was improved based on the YOLOv8n model. Specifically, the model introduces deeply separable convolution (DWConv) into the backbone network, integrates the large separated convolution kernel attention mechanism (LSKA) and Slim-Neck structure into the Neck part, and adopts the DIoU loss function for optimization. The experimental results show that on the oyster mushroom dataset, the OMC-YOLO model had a higher detection effect compared to mainstream target detection models such as Faster R-CNN, SSD, YOLOv3-tiny, YOLOv5n, YOLOv6, YOLOv7-tiny, YOLOv8n, YOLOv9-gelan, YOLOv10n, etc., and that the mAP50 value reached 94.95%, which is an improvement of 2.62%. The number of parameters and the computational amount were also reduced by 26%. The model provides technical support for the automatic detection of oyster mushroom grades, which helps in realizing quality control and reducing labor costs and has positive significance for the construction of smart agriculture.

Efficient Violence Recognition in Video Streams using ResDLCNN-GRU Attention Network

Detecting Violence in video streams is essential for public safety and security due to the rising frequency of violent incidents. Despite the extensive deployment of CCTV for surveillance, the available human monitoring resources still need to catch up with the need for vigilant supervision. This research presents a new lightweight model to address this gap by accurately identifying and categorizing violent behaviors in various scenarios, including CCTV footage. The proposed method leverages optical flow and RGB data to capture spatiotemporal features in the Violence data. Built on a Residual DLCNN architecture integrated with the Attention mechanism and GRU components, the model effectively handles high-dimensional video data, enhancing accuracy by prioritizing crucial frames containing violent and nonviolent instances. The proposed model's performance was validated on the Hockey Fights (HF), Movie Fights, and SCVD datasets, achieving impressive accuracies of 98.38%, 99.62%, and 90.57%, respectively. Here, we developed the Extended Automatic Violence Detection Dataset (EAVDD), featuring 1530 videos of violent scenes in movies, public spaces, social media, and sports. Testing the model with top fight scenes in rated movies yielded outstanding results. This research supports surveillance systems and advances short video analysis and understanding with applications in public safety, social media, sports, and law enforcement.

Lightweight bobbin yarn detection model for auto-coner with yarn bank

The automated replacement of empty tubes in the yarn bank is a critical step in the process of automatic winding machines with yarn banks, as the real-time detection of depleted yarn on spools and accurate positioning of empty tubes directly impact the production efficiency of winding machines. Addressing the shortcomings of traditional methods, such as poor adaptability and low sensitivity in optical and visual tube detection, and aiming to reduce the computational and detection time costs introduced by neural networks, this paper proposes a lightweight yarn spool detection model based on YOLOv8. The model utilizes Darknet-53 as the backbone network, and due to the dense spatial distribution of yarn spool targets, it incorporates large selective kernel units to enhance the recognition and positioning of dense targets. To address the issue of excessive focus on local features by convolutional neural networks, a bi-level routing attention mechanism is introduced to capture long-distance dependencies dynamically. Furthermore, to balance accuracy and detection speed, a FasterNeck is constructed as the neck network, replacing the original convolutional blocks with Ghost convolutions and integrating with FasterNet. This design minimizes the sacrifice of detection accuracy while achieving a significant improvement in inference speed. Lastly, the model employs weighted IoU with a dynamic focusing mechanism as the bounding box loss function. Experimental results on a custom yarn spool dataset demonstrate a notable improvement over the baseline model, with a high-confidence mAP of 94.2% and a compact weight size of only 4.9 MB. The detection speed reaches 223FPS, meeting the requirements for industrial deployment and real-time detection.

Lightweight model for small target detection of SAR images of ships based on NWD loss

Lightweight model for small target detection of SAR images of ships based on NWD loss

BAFormer: A Novel Boundary-Aware Compensation UNet-like Transformer for High-Resolution Cropland Extraction

Utilizing deep learning for semantic segmentation of cropland from remote sensing imagery has become a crucial technique in land surveys. Cropland is highly heterogeneous and fragmented, and existing methods often suffer from inaccurate boundary segmentation. This paper introduces a UNet-like boundary-aware compensation model (BAFormer). Cropland boundaries typically exhibit rapid transformations in pixel values and texture features, often appearing as high-frequency features in remote sensing images. To enhance the recognition of these high-frequency features as represented by cropland boundaries, the proposed BAFormer integrates a Feature Adaptive Mixer (FAM) and develops a Depthwise Large Kernel Multi-Layer Perceptron model (DWLK-MLP) to enrich the global and local cropland boundaries features separately. Specifically, FAM enhances the boundary-aware method by adaptively acquiring high-frequency features through convolution and self-attention advantages, while DWLK-MLP further supplements boundary position information using a large receptive field. The efficacy of BAFormer has been evaluated on datasets including Vaihingen, Potsdam, LoveDA, and Mapcup. It demonstrates high performance, achieving mIoU scores of 84.5%, 87.3%, 53.5%, and 83.1% on these datasets, respectively. Notably, BAFormer-T (lightweight model) surpasses other lightweight models on the Vaihingen dataset with scores of 91.3% F1 and 84.1% mIoU.

Lightweight deep learning model for automated STN localization using MER in Parkinson’s disease

Deep brain stimulation (DBS) of Subthalamic nucleus (STN), using a permanent electrode has emerged as an effective treatment for Parkinson’s disease patients. However, accurately localizing STN region during DBS surgery remains a challenging task. Microelectrode recording (MER) is used to identify STN by neurosurgeons. Nevertheless, manual interpretation of STN functional activity might prompt false alarms. Although the existing technology based on deep learning (DL) can outline the STN anatomical borders, it suffers from computational complexity and skewed findings. An end-to-end attention-based lightweight model with residual connections and soft thresholding is designed specifically for STN region detection. Scalogram and spectrogram images are generated from MER to mimic the observation of neurosurgeons and gather contextual information in a more intuitive manner. Proposed model exhibits a simplified architecture with only 15 layers, distinguishing it from previous DL models that tend to be more complex. Spatial attention helped to focus on relevant patterns within the data, while residual connections enhanced the gradient flow throughout the developed network. Soft thresholding offered noise reduction and robustness to signal variability. This approach eliminated the requirement for excessively deep layers and the necessity for separate feature extraction and classification stages by integrating them into a single pipeline. Our model identified STN with an accuracy of 97.42%. Proposed model outperformed 15 existing cutting-edge techniques. Our findings highlight the effectiveness of lightweight DL framework to achieve precise STN localization. The proposed method has significant potential in improving the efficiency of DBS surgery by reducing reliance on the neurosurgeon experience.

Open-Set Sheep Face Recognition in Multi-View Based on Li-SheepFaceNet

Deep learning-based sheep face recognition improves the efficiency and effectiveness of individual sheep recognition and provides technical support for the development of intelligent livestock farming. However, frequent changes within the flock and variations in facial features in different views significantly affect the practical application of sheep face recognition. In this study, we proposed the Li-SheepFaceNet, a method for open-set sheep face recognition in multi-view. Specifically, we employed the Seesaw block to construct a lightweight model called SheepFaceNet, which significantly improves both performance and efficiency. To enhance the convergence and performance of low-dimensional embedded feature learning, we used Li-ArcFace as the loss function. The Li-SheepFaceNet achieves an open-set recognition accuracy of 96.13% on a self-built dataset containing 3801 multi-view face images of 212 Ujumqin sheep, which surpasses other open-set sheep face recognition methods. To evaluate the robustness and generalization of our approach, we conducted performance testing on a publicly available dataset, achieving a recognition accuracy of 93.33%. Deploying Li-SheepFaceNet on an open-set sheep face recognition system enables the rapid and accurate identification of individual sheep, thereby accelerating the development of intelligent sheep farming.

CapNet: An Automatic Attention-Based with Mixer Model for Cardiovascular Magnetic Resonance Image Segmentation.

Deep neural networks have shown excellent performance in medical image segmentation, especially for cardiac images. Transformer-based models, though having advantages over convolutional neural networks due to the ability of long-range dependence learning, still have shortcomings such as having a large number of parameters and and high computational cost. Additionally, for better results, they are often pretrained on a larger data, thus requiring large memory size and increasing resource expenses. In this study, we propose a new lightweight but efficient model, namely CapNet, based on convolutions and mixing modules for cardiac segmentation from magnetic resonance images (MRI) that can be trained from scratch with a small amount of parameters. To handle varying sizes and shapes which often occur in cardiac systolic and diastolic phases, we propose attention modules for pooling, spatial, and channel information. We also propose a novel loss called the Tversky Shape Power Distance function based on the shape dissimilarity between labels and predictions that shows promising performances compared to other losses. Experiments on three public datasets including ACDC benchmark, Sunnybrook data, and MS-CMR challenge are conducted and compared with other state of the arts (SOTA). For binary segmentation, the proposed CapNet obtained the Dice similarity coefficient (DSC) of 94% and 95.93% for respectively the Endocardium and Epicardium regions with Sunnybrook dataset, 94.49% for Endocardium, and 96.82% for Epicardium with the ACDC data. Regarding the multiclass case, the average DSC by CapNet is 93.05% for the ACDC data; and the DSC scores for the MS-CMR are 94.59%, 92.22%, and 93.99% for respectively the bSSFP, T2-SPAIR, and LGE sequences of the MS-CMR. Moreover, the statistical significance analysis tests with p-value compared with transformer-based methods and some CNN-based approaches demonstrated that the CapNet, though having fewer training parameters, is statistically significant. The promising evaluation metrics show comparative results in both Dice and IoU indices compared to SOTA CNN-based and Transformer-based architectures.

Lightweight-detection: The strip steel surface defect identification based on improved YOLOv5d

The surface defect of strip steel seriously affects the product quality, the current identification methods have the inherent defects of low detection efficiency and poor detection accuracy, and it is important an improved an algorithm based on the modified YOLOv5 framework to enhance the defect detection effectiveness and accuracy. To create a more lightweight network model, the C3 module and part of the convolutional structure in the original YOLOv5 network were replaced with the GhostBottleneck structure. Additionally, the Coordinate Attention (CA) mechanism was incorporated into the Backbone section to compensate for the original model's lack of attention mechanisms. To address the angular mismatch between the actual frame and the predicted bounding box, a limitation overlooked by the CIOU loss function was further refined. Experimental results demonstrate that the enhanced YOLOv5s-CSG model outperformed the original YOLOv5s by 7.3 %, achieving a significant 37.9 % reduction in model size and an mAP value of 77.5 %. Furthermore, the detection precision and speed were notably higher compared to several widely-used algorithms. The proposed YOLOv5s-CSG model is well-suited for deployment on mobile devices and shows great potential for rapidly and accurately detecting both the type and location of strip surface defects, surpassing the performance of existing methods.

Gesture Recognition of Filipino Sign Language Using Convolutional and Long Short-Term Memory Deep Neural Networks

In response to the recent formalization of Filipino Sign Language (FSL) and the lack of comprehensive studies, this paper introduces a real-time FSL gesture recognition system. Unlike existing systems, which are often limited to static signs and asynchronous recognition, it offers dynamic gesture capturing and recognition of 10 common expressions and five transactional inquiries. To this end, the system sequentially employs cropping, contrast adjustment, grayscale conversion, resizing, and normalization of input image streams. These steps serve to extract the region of interest, reduce the computational load, ensure uniform input size, and maintain consistent pixel value distribution. Subsequently, a Convolutional Neural Network and Long-Short Term Memory (CNN-LSTM) model was employed to recognize nuances of real-time FSL gestures. The results demonstrate the superiority of the proposed technique over existing FSL recognition systems, achieving an impressive average accuracy, recall, and precision rate of 98%, marking an 11.3% improvement in accuracy. Furthermore, this article also explores lightweight conversion methods, including post-quantization and quantization-aware training, to facilitate the deployment of the model on resource-constrained platforms. The lightweight models show a significant reduction in model size and memory utilization with respect to the base model when executed in a Raspberry Pi minicomputer. Lastly, the lightweight model trained with the quantization-aware technique (99%) outperforms the post-quantization approach (97%), showing a notable 2% improvement in accuracy.

Lightweight Low-Voltage AC Arc-Fault Detection Method Based on the Interpretability Method

Electrical fires are frequently caused by low-voltage AC series arc faults, which can result in significant injuries and property damage. The installation of arc-fault detection devices is mandated or recommended in many regions and countries across the world, yet the current devices’ detection accuracy is insufficient to completely eliminate the risk posed by arc faults. The method based on artificial intelligence is a solution with high detection accuracy, but the AI model is a ‘black box’. When a misjudgment occurs, the cause of the model error cannot be found fundamentally, and the modification and light weight of the model also presents significant difficulties when using the approach. Given the aforementioned issues, this research proposes a novel lightweight low-voltage AC arc-fault detection method based on the explainability approach. By applying the attention mechanism approach and performing a visual analysis, the contribution of arc features to model detection is determined. Model input data optimization and model structure simplification are achieved at the same time as increased model detection accuracy. Ultimately, an experimental prototype for arc-fault detection is designed and validated. Test results demonstrate the effectiveness of the method by demonstrating that the lightweight model maintains 99.69% detection accuracy, even after optimizing the input data by 80% and reducing the model parameters by 51.52%.

MC-ShuffleNetV2: A lightweight model for maize disease recognition

Maize has a long history of cultivation and is renowned for its high yield, superior quality, and adaptability. Currently, maize holds a significant position in grain cultivation and occupies a significant place in the agricultural structure. However, maize is susceptible to various diseases during its growth process, which can have a significant impact on the quality and yield. Traditional machine learning is heavily reliant on feature extraction, whereas deep learning has demonstrated notable success in image recognition for computer vision.The use of bloated models and the resulting wastage of computational resources represent significant challenges. The paper proposes a lightweight model, MC-ShuffleNetV2 (Mish + Convolutional Block Attention Module + ShuffleNetV2), to meet the practical needs of convolutional neural networks in maize disease image recognition. The model has designed with a focus on network lightweighting and accurate feature extraction. The model was constructed upon the foundation of the high-performance ShuffleNetV2 1 × network. The Convolutional Block Attention Module was integrated into the network architecture to enhance the model’s adaptive expressiveness. The depthwise separable convolution kernel of the depth-separable module was modified from a 3 × 3 kernel to a 5 × 5 kernel. This modification was implemented with the objective of expanding the image receptive field and extracting more detailed features of the image. It was necessary to modify the activation function in each stage for Mish. The model was compressed through the application of pruning operations. In the maize disease dataset test, the accuracy of the test set recognition accuracy of the network model constructed in this paper reaches 99.86 %, the model parameters are only 873,936, and the FLOPs (Floating-point Operations) are only 1,751,286. Compared with LeNet, AlexNet, MobileNetV2, and EfficientNetV2 models, the MC-ShufflenetV2 model’s recognition ability and size have obvious advantages, and it is more conducive to the actual deployment of the agricultural mobile terminal.

Efficient and lightweight layer-wise in-situ defect detection in laser powder bed fusion via knowledge distillation and structural re-parameterization

Powder bed quality is critical in Laser Powder Bed Fusion (LPBF) as defects in the powder bed will affect the quality of the forming parts. Recently, numerous powder bed defect detection methods based on semantic segmentation algorithms have been developed. However, due to the computational resource constraints and real-time requirements in industry scenarios, these heavy models face challenges in deploying into the LPBF process monitoring system. Meanwhile, existing lightweight models struggle to achieve sufficient performance on industrial data due to their limited feature extraction capabilities. To address the challenges in industrial deployment, this paper proposes a lightweight and efficient LPBF layer-wise defect detection paradigm to achieve a balance between performance and efficiency. Aiming at improving the performance of the lightweight model, this paper proposes a structured Knowledge Distillation framework, in which a pre-trained high-performance segmentation model (UNet++) is used to guide the training process of a lightweight model. To address the impact of model discrepancies on the effectiveness of Knowledge Distillation and further improve the inference speed, this paper constructs a specific lightweight model and applies Structural Re-parameterization to deeply compress the model. Moreover, we created a dataset comprising 406 images of powder-bed defects, with each image annotated at the pixel level. Massive experiments based on the dataset demonstrate the effectiveness of the proposed paradigm, with an excellent trade-off in performance and efficiency. The speed of the proposed model is approximately 5 times faster than that of high-performance models while maintaining comparable performance.

PLM-Res-U-Net: A light weight binarization model for enhancement of multi-textured palm leaf manuscript images

This paper proposes a deep semantic binarization model, PLM-Res-U-Net, for enhancing palm-leaf manuscripts. PLM-Res-U-Net is a lightweight model comprising encoding and decoding blocks with skip connections. The model enhances the palm leaf manuscript by efficiently retaining the text strokes by removing the degradations such as uneven illumination, aging marks, brittleness, and background discolorations. Two datasets of palm leaf manuscript collections with multiple degradation patterns and diverse textured backgrounds are used for experimentation. PLM-Res-U-Net is trained from scratch with 50 epochs with a learning rate of1e−8 with three sampling strategies. The performance of state-of-the-art deep learning models ResUnet, Pspnet, U-Net++, and Segnet are also evaluated along with two diverse benchmark datasets. Analysis shows that results obtained by the proposed PLM-Res-U-Net prove generalizability and computational efficacy with a dice score of 0.986. Additionally, PLM-Res-U-Net successfully preserves the edge strokes of the text compared with state-of-the-art models.

Research on prevention and control tracking system based on 3D face recognition

In order to improve the tracking effect of prevention and control, this paper combines three-dimensional face recognition technology and posture recognition method to build a prevention and control tracking system, and designs a single-stage human target detection and posture estimation network based on thermal infrared images. By using infrared recognition methods, the face and human body shape can be directly recognized, overcoming the problem of traditional visual recognition methods being easily occluded. The network can simultaneously complete two tasks of human target detection and pose estimation in a multi-task manner. Moreover, this paper uses the knowledge distillation strategy to train a lightweight model to further reduce the amount of model parameters to improve the inference speed. In addition, this paper uses a single-stage human target detection and posture estimation network to judge the effect of prevention and control tracking. Through the prevention and control tracking effect test, it can be seen that the prevention and control tracking system based on 3D face recognition proposed in this paper can effectively improve the prevention and control tracking effect.

Derivative Method to Detect Sleep and Awake States through Heart Rate Variability Analysis Using Machine Learning Algorithms.

Sleep disorders can have harmful consequences in both the short and long term. They can lead to attention deficits, as well as cardiac, neurological and behavioral repercussions. One of the most widely used methods for assessing sleep disorders is polysomnography (PSG). A major challenge associated with this method is all the cables needed to connect the recording devices, making the examination more intrusive and usually requiring a clinical environment. This can have potential consequences on the test results and their accuracy. One simple way to assess the state of the central nervous system (CNS), a well-known indicator of sleep disorder, could be the use of a portable medical device. With this in mind, we implemented a simple model using both the RR interval (RRI) and its second derivative to accurately predict the awake and napping states of a subject using a feature classification model. For training and validation, we used a database providing measurements from nine healthy young adults (six men and three women), in which heart rate variability (HRV) associated with light-on, light-off, sleep onset and sleep offset events. Results show that using a 30 min RRI time series window suffices for this lightweight model to accurately predict whether the patient was awake or napping.

ALdamage-seg: A Lightweight Model for Instance Segmentation of Aluminum Profiles

Aluminum profiles are widely used in various manufacturing sectors due to their flexibility and chemical properties. However, these profiles are susceptible to defects during manufacturing and transportation. Detecting these defects is crucial, but existing object detection models like Mask R-CNN and YOLOv8-seg are not optimized for this task. These models are large and computationally intensive, making them unsuitable for edge devices used in industrial inspections. To address this issue, this study proposes a novel lightweight instance segmentation model called AL-damage-seg, inspired by the YOLOv8n-seg architecture. This model utilizes MobileNetV3 as the backbone. In YOLOv8n-seg, the role of C2f is to enhance the nonlinear representation of the model to capture complex image features more efficiently. We upgraded and improved it to form multilayer feature extraction module (MFEM) and integrates a large separable kernel attention (LSKA) mechanism in the C2f module, resulting in C2f_LSKA, to further optimize the performance of the model. Additionally, depth-wise separable convolutions are employed in the feature fusion process. The ALdamage-seg’s weight on the Alibaba Tian-chi aluminum profile dataset constitutes 43.9% of that of YOLOv8n-seg, with its GFLOPs reduced to 53% relative to YOLOv8-seg, all the while achieving an average precision (mAP) of 99% relative to YOLOv8-seg. With its compact size and lower computational requirements, this model is well-suited for deployment on edge devices with limited processing capabilities.

Leverage Effective Deep Learning Searching Method for Forensic Age Estimation.

Dental age estimation is extensively employed in forensic medicine practice. However, the accuracy of conventional methods fails to satisfy the need for precision, particularly when estimating the age of adults. Herein, we propose an approach for age estimation utilizing orthopantomograms (OPGs). We propose a new dental dataset comprising OPGs of 27,957 individuals (16,383 females and 11,574 males), covering an age range from newborn to 93 years. The age annotations were meticulously verified using ID card details. Considering the distinct nature of dental data, we analyzed various neural network components to accurately estimate age, such as optimal network depth, convolution kernel size, multi-branch architecture, and early layer feature reuse. Building upon the exploration of distinctive characteristics, we further employed the widely recognized method to identify models for dental age prediction. Consequently, we discovered two sets of models: one exhibiting superior performance, and the other being lightweight. The proposed approaches, namely AGENet and AGE-SPOS, demonstrated remarkable superiority and effectiveness in our experimental results. The proposed models, AGENet and AGE-SPOS, showed exceptional effectiveness in our experiments. AGENet outperformed other CNN models significantly by achieving outstanding results. Compared to Inception-v4, with the mean absolute error (MAE) of 1.70 and 20.46 B FLOPs, our AGENet reduced the FLOPs by 2.7×. The lightweight model, AGE-SPOS, achieved an MAE of 1.80 years with only 0.95 B FLOPs, surpassing MobileNetV2 by 0.18 years while utilizing fewer computational operations. In summary, we employed an effective DNN searching method for forensic age estimation, and our methodology and findings hold significant implications for age estimation with oral imaging.

Unveiling suspicious phishing attacks: enhancing detection with an optimal feature vectorization algorithm and supervised machine learning

IntroductionThe dynamic and sophisticated nature of phishing attacks, coupled with the relatively weak anti-phishing tools, has made phishing detection a pressing challenge. In light of this, new gaps have emerged in phishing detection, including the challenges and pitfalls of existing phishing detection techniques. To bridge these gaps, this study aims to develop a more robust, effective, sophisticated, and reliable solution for phishing detection through the optimal feature vectorization algorithm (OFVA) and supervised machine learning (SML) classifiers.MethodsInitially, the OFVA was utilized to extract the 41 optimal intra-URL features from a novel large dataset comprising 2,74,446 raw URLs (134,500 phishing and 139,946 legitimate URLs). Subsequently, data cleansing, curation, and dimensionality reduction were performed to remove outliers, handle missing values, and exclude less predictive features. To identify the optimal model, the study evaluated and compared 15 SML algorithms arising from different machine learning (ML) families, including Bayesian, nearest-neighbors, decision trees, neural networks, quadratic discriminant analysis, logistic regression, bagging, boosting, random forests, and ensembles. The evaluation was performed based on various metrics such as confusion matrix, accuracy, precision, recall, F-1 score, ROC curve, and precision-recall curve analysis. Furthermore, hyperparameter tuning (using Grid-search) and k-fold cross-validation were performed to optimize the detection accuracy.Results and discussionThe findings indicate that random forests (RF) outperformed the other classifiers, achieving a greater accuracy rate of 97.52%, followed by 97.50% precision, and an AUC value of 97%. Finally, a more robust and lightweight anti-phishing model was introduced, which can serve as an effective tool for security experts, practitioners, and policymakers to combat phishing attacks.

EMG-YOLO: road crack detection algorithm for edge computing devices.

Road cracks significantly shorten the service life of roads. Manual detection methods are inefficient and costly. The YOLOv5 model has made some progress in road crack detection. However, issues arise when deployed on edge computing devices. The main problem is that edge computing devices are directly connected to sensors. This results in the collection of noisy, poor-quality data. This problem adds computational burden to the model, potentially impacting its accuracy. To address these issues, this paper proposes a novel road crack detection algorithm named EMG-YOLO. First, an Efficient Decoupled Header is introduced in YOLOv5 to optimize the head structure. This approach separates the classification task from the localization task. Each task can then focus on learning its most relevant features. This significantly reduces the model's computational resources and time. It also achieves faster convergence rates. Second, the IOU loss function in the model is upgraded to the MPDIOU loss function. This function works by minimizing the top-left and bottom-right point distances between the predicted bounding box and the actual labeled bounding box. The MPDIOU loss function addresses the complex computation and high computational burden of the current YOLOv5 model. Finally, the GCC3 module replaces the traditional convolution. It performs global context modeling with the input feature map to obtain global context information. This enhances the model's detection capabilities on edge computing devices. Experimental results show that the improved model has better performance in all parameter indicators compared to current mainstream algorithms. The EMG-YOLO model improves the accuracy of the YOLOv5 model by 2.7%. The mAP (0.5) and mAP (0.9) are improved by 2.9% and 0.9%, respectively. The new algorithm also outperforms the YOLOv5 model in complex environments on edge computing devices. The EMG-YOLO algorithm proposed in this paper effectively addresses the issues of poor data quality and high computational burden on edge computing devices. This is achieved through optimizing the model head structure, upgrading the loss function, and introducing global context modeling. Experimental results demonstrate significant improvements in both accuracy and efficiency, especially in complex environments. Future research can further optimize this algorithm and explore more lightweight and efficient object detection models for edge computing devices.