Search

It is crucial to ensure workers wear safety helmets when working at a workplace with a high risk of safety accidents, such as construction sites and mine tunnels. Although existing methods can achieve helmet detection in images, their accuracy and speed still need improvements since complex, cluttered, and large-scale scenes of real workplaces cause server occlusion, illumination change, scale variation, and perspective distortion. So, a new safety helmet-wearing detection method based on deep learning is proposed. Firstly, a new multi-scale contextual aggregation module is proposed to aggregate multi-scale feature information globally and highlight the details of concerned objects in the backbone part of the deep neural network. Secondly, a new detection block combining the dilate convolution and attention mechanism is proposed and introduced into the prediction part. This block can effectively extract deep features while retaining information on fine-grained details, such as edges and small objects. Moreover, some newly emerged modules are incorporated into the proposed network to improve safety helmet-wearing detection performance further. Extensive experiments on open dataset validate the proposed method. It reaches better performance on helmet-wearing detection and even outperforms the state-of-the-art method. To be more specific, the mAP increases by 3.4%, and the speed increases from 17 to 33 fps in comparison with the baseline, You Only Look Once (YOLO) version 5X, and the mean average precision increases by 1.0% and the speed increases by 7 fps in comparison with the YOLO version 7. The generalization ability and portability experiment results show that the proposed improvements could serve as a springboard for deep neural network design to improve object detection performance in complex scenarios. [ABSTRACT FROM AUTHOR]

Abstract In the field of industrial safety, wearing helmets plays a vital role in ensuring workers’ health. Aiming at addressing the complex background in the industrial environment, caused by differences in distance, the helmet small target wearing detection methods for misdetection and omission detection problems are needed. An improved YOLOv8 safety helmet wearing detection network is proposed to enhance the capture of details, improve multiscale feature processing and improve the accuracy of small target detection by introducing Dilation-wise residual attention module, atrous spatial pyramid pooling and normalized Wasserstein distance loss function. Experiments were conducted on the SHWD dataset, and the results showed that the mAP of the improved network improved to 92.0%, which exceeded that of the traditional target detection network in terms of accuracy, recall, and other key metrics. These findings further improved the detection of helmet wearing in complex environments and greatly enhanced the accuracy of detection.

Safety helmet wearing detection based on video surveillance is an important means of safety monitoring in many industrial scenes. The training of safety helmet wearing detection models requires large and well-labeled dataset. However, the incidence of security violations is relatively low, which results in insufficient samples for training deep detection models. Safety helmet wearing detection is a common requirement in many scenarios such as construction sites, substations, and factory workshops. Aggregating data from multiple companies for model training would improve the performance of the detection model. Traditional centralized training methods are not feasible because aggregating data in centralized locations (such as the cloud) can raise concerns about data privacy and the high cost of data communication and storage. This paper proposes FedSH, a novel cloud-edge-based federated learning framework, which learns a shared global safety helmet wearing detection model in the cloud from multiple companies at the network edges and achieves data privacy protection by keeping company data locally. In addition, this paper designs reweighting mechanisms and applies transfer learning to address class imbalance and non-IID problems in the training data, so as to obtain an accurate and personalized detection model. Extensive experiments have been conducted on real surveillance video datasets. The experimental results demonstrate that FedSH outperforms the existing widely used federated learning methods with an accuracy improvement of at least 3.4%; the reduction in accuracy is within the range of 5% compared with centralized learning methods. FedSH effectively achieves a good balance between model performance, privacy protection, and communication efficiency. [ABSTRACT FROM AUTHOR]

The existing algorithm for detecting the helmet wear by coal miners has the problem of difficulty in achieving a good balance between detection accuracy and speed. In order to solve the above problem, based on the YOLOv4 model, a lightweight model (M-YOLO) that integrates coordinate attention and multi-scale is proposed and applied in safety helmet wearing detection. This model replaces YOLOv4's feature extraction network CSPDarknet53 with a lightweight feature extraction network S-MobileNetV2 composed of a mixed coordinate attention module. It effectively improves the connection between features while reducing the number of related parameters. The model changes the parallel connection method in the original spatial pyramid pooling structure to serial connection. It effectively improves computational efficiency. The feature fusion network is improved by introducing shallow features with high-resolution and multi detail texture information. It effectively enhances the extraction of object features. Some convolutions in the original Neck structure are modified to deep separable convolutions, further reducing the model's parameter and computational complexity while ensuring detection precision. The experimental results show that compared with the YOLOv4 model, the mean average precision of the M-YOLO model is only reduced by 0.84%. But the computational complexity, parameter quantity, and model size are reduced by 74.5%, 72.8%, and 81.6%, respectively. The detection speed is improved by 53.4%. Compared to other models, the M-YOLO model achieves a good balance between accuracy and real-time performance, meeting the requirements of embedded loading and deployment on intelligent video surveillance terminals.

Construction workers protect themselves from accidental injury by wearing safety helmets correctly. The safety helmet wearing detection algorithms currently based on deep learning aim to solve the problem of whether the safety helmet is worn in the head region, rather than whether the safety helmet is worn properly. This paper presents a novel Spatial Position Relation Capsule Network, termed SPR-CapsNet, for addressing safety helmet wearing correctly detection. In the first step, the learned deep feature maps are divided into patches, and each patch is transformed into a vector as a primary capsule, which effectively reduces the computational cost. In the second step, the dynamic routing algorithm is used to learn the spatial relationship between local image features. In the final step, the decision-making is optimized based on the probability of different dimensions of the output vector. It will be recognized as wearing status when the safety helmet is in a more appropriate position relative to the face. SPR-CapsNet is compared with the original Capsule Network, Convolutional Neural Network (CNN), Vision Transformer (ViT), and other models on a large amount of data. Experiments demonstrate the superiority of the proposed network. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Mine Automation is the property of Industry & Mine Automation Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Establishing a lightweight yet high-precision object detection algorithm is paramount for accurately assessing workers’ helmet-wearing status in intricate industrial settings. Helmet detection is inherently challenging due to factors like the diminutive target size, intricate backgrounds, and the need to strike a balance between model compactness and detection accuracy. In this paper, we propose YOLO-LHD (You Only Look Once-Lightweight Helmet Detection), an efficient framework built upon the YOLOv8 object detection model. The proposed approach enhances the model’s ability to detect small targets in complex scenes by incorporating the Coordinate attention mechanism and Focal loss function, which introduce high-resolution features and large-scale detection heads. Additionally, we integrate the improved Ghostv2 module into the backbone feature extraction network to further improve the balance between model accuracy and size. We evaluated our method on MHWD dataset established in this study and compared it with the baseline model YOLOv8n. The proposed YOLO-LHD model achieved a reduction of 66.1% in model size while attaining the best 94.3% mAP50 with only 0.86M parameters. This demonstrates the effectiveness of the proposed approach in achieving lightweight deployment and high-precision helmet detection.

Safety helmet-wearing detection is an essential part of the intelligent monitoring system. To improve the speed and accuracy of detection, especially small targets and occluded objects, it presents a novel and efficient detector model. The underlying core algorithm of this model adopts the YOLOv5 (You Only Look Once version 5) network with the best comprehensive detection performance. It is improved by adding an attention mechanism, a CIoU (Complete Intersection Over Union) Loss function, and the Mish activation function. First, it applies the attention mechanism in the feature extraction. The network can learn the weight of each channel independently and enhance the information dissemination between features. Second, it adopts CIoU loss function to achieve accurate bounding box regression. Third, it utilizes Mish activation function to improve detection accuracy and generalization ability. It builds a safety helmet-wearing detection data set containing more than 10,000 images collected from the Internet for preprocessing. On the self-made helmet wearing test data set, the average accuracy of the helmet detection of the proposed algorithm is 96.7%, which is 1.9% higher than that of the YOLOv5 algorithm. It meets the accuracy requirements of the helmet-wearing detection under construction scenarios. [ABSTRACT FROM AUTHOR]

Wearing safety helmets can effectively protect workers safety on construction sites. However, workers often take off the helmets because of weak security-conscious and discomfort, then hidden dangers will be brought by this behaviour. Workers without safety helmets will suffer more injuries in accidents such as falling human body and vertical falling matter. Hence, detecting safety helmet wearing is a vital step of construction sites safety management and a safety helmet detector with high speed and accuracy is urgently needed. However, traditional manual monitor is labour intensive and methods of installing sensors on safety helmet are difficult to popularize. Therefore, this paper proposes a deep learning-based method to detect safety helmet wearing at a satisfactory accuracy with high detection speed. Our method chooses YOLO v5 as the baseline, then the fourth detection scale is added to predict more bounding boxes for small objects and the attention mechanism is adopted in the backbone of the network to construct more informative features for following concatenation operations. In order to overcome the defects caused by insufficient data, targeted data augmentation and transfer learning are used. Improvements caused by every modification are discussed in this paper. Finally, our model achieves 92.2% mean average precision, up 6.3% compared to the original algorithm, and it only takes 3.0 ms to detect an image at $640 \times 640$ . These results demonstrate the robustness and feasibility of our model. Meanwhile, the size of our trained model is only 16.3 m, which means the model is easy to be deployed. At last, after obtaining a satisfactory model, a graphical user interface (GUI) is designed to make our algorithm more user-friendly.

In order to overcome the problems of object detection in complex scenes based on the YOLOv4-tiny algorithm, such as insufficient feature extraction, low accuracy, and low recall rate, an improved YOLOv4-tiny safety helmet-wearing detection algorithm SCM-YOLO is proposed. Firstly, the Spatial Pyramid Pooling (SPP) structure is added after the backbone network of the YOLOv4-tiny model to improve its adaptability of different scale features and increase its effective features extraction capability. Secondly, Convolutional Block Attention Module (CBAM), Mish activation function, K-Means++ clustering algorithm, label smoothing, and Mosaic data enhancement are introduced to improve the detection accuracy of small objects while ensuring the detection speed. After a large number of experiments, the proposed SCM-YOLO algorithm achieves a mAP of 93.19%, which is 4.76% higher than the YOLOv4-tiny algorithm. Its inference speed reaches 22.9FPS (GeForce GTX 1050Ti), which meets the needs of the real-time and accurate detection of safety helmets in complex scenes. [ABSTRACT FROM AUTHOR]

Safety production management is an important policy for the development of high-risk enterprises such as the construction industry and heavy industry,and safety helmets play a key role in head protection in the production environment.Therefore,it is necessary to strengthen the supervision of helmet wearing.In recent years,the monitoring method of helmet wearing based on image vision has become the main means for enterprises to implement management.How to improve the detection accuracy and speed of helmet wearing is a crucial issue for applications.To deal with this issue,an improved YOLO v4 algorithm is proposed to promote the accuracy and efficiency of safety helmet wearing detection in this paper.First,a 128×128 feature map output is added to the original three feature map outputs of the YOLO v4 algorithm,and the 8 times downsampling of the feature map output is changed to 4 times downsampling to provide more small target features for subsequent feature fusion.Second,the feature fusion module is improved based on the idea of dense connection to realize feature reuse,so that the Yolo Head classifier responsible for small target detection can utilize the features of different levels,to obtain better target detection and classification results.Finally,comparative experiments are carried out.The results show that the average accuracy of the proposed method is 2.96% higher than the original network detection accuracy to be 91.17%,and the detection speed is basically unchanged to be 52.9 frame/s.Thereupon,the proposed algorithm can achieve better detection accuracy while meeting real-time detection requirements,and effectively realize the high-speed and high-precision detection of helmet wearing.

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Automatic detection of workers wearing safety helmets at the construction site is essential for safe production. Aiming at the problem of low recognition rate caused by factors such as background and light in the automatic detection of safety helmets using traditional machine learning methods, this paper proposes an object detection framework that combines Online Hard Example Mining (OHEM) and multi-part combination. In our framework, we first use the multi-scale training and the increasing anchors strategies to enhance the robustness of the original Faster RCNN algorithm to detect different scales and small object. Then, the OHEM is to optimize the model to prevent the imbalance of positive and negative samples. Finally, the person wearing the helmet and its parts (helmet and person) are detected by improved Faster RCNN. The multi-part combination method uses the geometric information of the detection objects to determine if a worker is wearing a helmet. Experiments show that compared with the original Faster RCNN, the detection accuracy is increased by 7%. It also has better detection performance for partial occlusion and different-size objects, showing good generalization and robustness.

Concerning the problem that the existing safety helmet wearing detection method has certain detection difficulty and low accuracy for partial occlusion, different size and small object, this paper proposed a safety helmet wearing detection method based on improved Faster RCNN and multi-part model. Based on original Faster RCNN algorithm, this method used multi-scale training and increased anchor number strategy to enhance the robustness of network to detect different scales, and introduced online hard example mining strategy to prevent imbalanced between positive and negative samples . Then, it detected the wearing safety helmet workers, safety helmet and other parts. Last, it used multi-part model method to eliminate the false objects. Experiments show that the detection accuracy is improved by 7% compared to the original Faster RCNN, and the adaptability to the environment is stronger. [ABSTRACT FROM AUTHOR]