Knowledge, Attitude, and Practice of Using Helmets in Children amongst Parents to Prevent Head Injuries: A Cross-Sectional Study in Riyadh, Saudi Arabia

Objectives: The primary objective was to characterize the range of Knowledge, Attitude, and Practice (KAP) of Helmet use in children amongst parents to prevent head injuries and death. Methods: This is a cross-sectional study, done by online survey using a snowball sampling technique, the number of included responses were 386 parents (Male and female) living in Riyadh Aged 21 - 60 years old or above. Results: The study showed that there is a difference in Parents’ belief in the importance of helmet use while riding a Bicycle vs Motorcycle/Quad bike and that was affected by parents’ education level, almost all the people who answered the survey (76.7%) agree that it is important for their children to wear a helmet when riding both a Bicycle and a Motorcycle or Quadbike with a cumulative percentage of (93.8%). And almost all agreed on multiple approaches to help increase helmet use be it by forcing rental shops to give out helmets, forcing sellers to recommend the use of helmets, increasing awareness campaigns, and imposing fines for not wearing helmets. Conclusions: This study is the first to explore Family helmet use while riding Bicycles and Motorcycles/Quad bikes. Although Parent’s belief in the importance of helmet use for their children was high, it is clear that the level of practice is low. With that the risk of head injuries might be high, our findings suggest that safety interventions for increasing pediatric helmet use are needed to increase helmet use and reduce the risk of head injury and hospitalization.

Improved YOLOv8n Model for Detecting Helmets and License Plates on Electric Bicycles

Wearing helmetswhile riding electric bicycles can significantly reduce head injuries resulting fromtraffic accidents.To effectively monitor compliance,the utilization of target detection algorithms through traffic cameras plays a vital role in identifying helmet usage by electric bicycle riders and recognizing license plates on electric bicycles.However,manual enforcement by traffic police is time-consuming and labor-intensive.Traditional methods face challenges in accurately identifying small targets such as helmets and license plates using deep learning techniques.This paper proposes an enhanced model for detecting helmets and license plates on electric bicycles,addressing these challenges.The proposedmodel improves uponYOLOv8n by deepening the network structure,incorporating weighted connections,and introducing lightweight convolutional modules.These modifications aim to enhance the precision of small target recognition while reducing the model’s parameters,making it suitable for deployment on low-performance devices in real traffic scenarios.Experimental results demonstrate that the model achieves an mAP@0.5 of 91.8%,showing an 11.5%improvement over the baselinemodel,with a 16.2%reduction in parameters.Additionally,themodel achieves a frames per second(FPS)rate of 58,meeting the accuracy and speed requirements for detection in actual traffic scenarios.

HWD-YOLO:A New Vision-Based Helmet Wearing Detection Method

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 featureswhile 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 helmetwearing 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 from17 to 33 fps in comparison with the baseline,You Only Look Once(YOLO)version 5X,and themean 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.

Modelling ballistic impact on military helmets:The relevance of projectile plasticity

The need to develop armour systems to protect against attacks from various sources is increasingly a matter of personal,social and national security.To develop innovative armour systems it is necessary to monitor developments being made on the type,technology and performance of the threats(weapons,projectiles,explosives,etc.) Specifically,the use of high protection level helmets on the battlefield is essential.The development of evaluation methods that can predict injuries and trauma is therefore of major importance.However,the risk of injuries or trauma that can arise from induced accelerations is an additional consideration.To develop new materials and layouts for helmets it is necessary to study the effects caused by ballistic impacts in the human head on various scenarios.The use of numerical simulation is a fundamental tool in this process.The work here presented focuses on the use of numerical simulation(finite elements analysis) to predict the consequences of bullet impacts on military helmets on human injuries.The main objectives are to assess the level and probability of head trauma using the Head Injury Criterion,caused by the impact of a 9 mm NATO projectile on a PASGT helmet and to quantify the relevance of projectile plasticity on the whole modelling process.The accelerations derived from the impact phenomenon and the deformations caused on the helmet are evaluated using fully three-dimensional models of the helmet,head,neck and projectile.Impact studies are done at impact angles ranging from 0 to 75°.Results are presented and discussed in terms of HIC and probability of acceleration induced trauma levels.Thorough comparison analyses are done using a rigid and a deformable projectile and it is observed that plastic deformation of the projectile is a significant energy dissipation mechanism in the whole impact process.

Detection of Worker’s Safety Helmet and Mask and Identification of Worker Using Deeplearning

This paper proposes a method for detecting a helmet for thesafety of workers from risk factors and a mask worn indoors and verifying aworker’s identity while wearing a helmet and mask for security. The proposedmethod consists of a part for detecting the worker’s helmet and mask and apart for verifying the worker’s identity. An algorithm for helmet and maskdetection is generated by transfer learning of Yolov5’s s-model and m-model.Both models are trained by changing the learning rate, batch size, and epoch.The model with the best performance is selected as the model for detectingmasks and helmets. At a learning rate of 0.001, a batch size of 32, and anepoch of 200, the s-model showed the best performance with a mAP of0.954, and this was selected as an optimal model. The worker’s identificationalgorithm consists of a facial feature extraction part and a classifier partfor the worker’s identification. The algorithm for facial feature extraction isgenerated by transfer learning of Facenet, and SVMis used as the classifier foridentification. The proposed method makes trained models using two datasets,a masked face dataset with only a masked face, and a mixed face datasetwith both a masked face and an unmasked face. And the model with the bestperformance among the trained models was selected as the optimal model foridentification when using a mask. As a result of the experiment, the model bytransfer learning of Facenet and SVM using a mixed face dataset showed thebest performance. When the optimal model was tested with a mixed dataset,it showed an accuracy of 95.4%. Also, the proposed model was evaluated asdata from 500 images of taking 10 people with a mobile phone. The resultsshowed that the helmet and mask were detected well and identification wasalso good.

Detection of Safety Helmet-Wearing Based on the YOLO_CA Model

Safety helmets can reduce head injuries from object impacts and lower the probability of safety accidents,as well as being of great significance to construction safety.However,for a variety of reasons,construction workers nowadays may not strictly enforce the rules of wearing safety helmets.In order to strengthen the safety of construction site,the traditional practice is to manage it through methods such as regular inspections by safety officers,but the cost is high and the effect is poor.With the popularization and application of construction site video monitoring,manual video monitoring has been realized for management,but the monitors need to be on duty at all times,and thus are prone to negligence.Therefore,this study establishes a lightweight model YOLO_CA based on YOLOv5 for the automatic detection of construction workers’helmet wearing,which overcomes the shortcomings of the current manual monitoring methods that are inefficient and expensive.The coordinate attention(CA)addition to the YOLOv5 backbone strengthens detection accuracy in complex scenes by extracting critical information and suppressing non-critical information.Further parameter compression with deeply separable convolution(DWConv).In addition,to improve the feature representation speed,we swap out C3 with a Ghost module,which decreases the floating-point operations needed for feature channel fusion,and CIOU_Loss was substituted with EIOU_Loss to enhance the algorithm’s localization accuracy.Therefore,the original model needs to be improved so as to enhance the detection of safety helmets.The experimental results show that the YOLO_CA model achieves good results in all indicators compared with the mainstream model.Compared with the original model,the mAP value of the optimized model increased by 1.13%,GFLOPs cut down by 17.5%,and there is a 6.84%decrease in the total model parameters,furthermore,the weight size cuts down by 4.26%,FPS increased by 39.58%,and the detection effect and model size of this model can meet the requirements of lightweight embedding.

Preventable head and facial injuries by providing free bicycle helmets and education to preschool children in a head start program

The objectives of the study were to determine helmet use rates, incidence rates (IRs) of head and facial injuries for population attributable fraction (PAF) estimation, and to elucidate the magnitude of and changes in PAFs as the result of helmet use changes among preschool children. A study consisting of cross-sectional (survey) and longitudinal (follow-up) component was designed by including a randomly selected group of participants (n = 322) from 10 Head Start sites provided with free bicycle helmets along with a subgroup of prior helmet owners (n = 68) from the other random group (n = 285). All participants received bicycle helmet education. Helmet use surveys were conducted in May (1st Survey) and November 2008 (2nd Survey). The helmet owners were followed up to determine IRs, and incidence rate ratios (IRRs) for head and facial injuries. PAFs were computed using IRs as well as helmet use rates and IRRs. Helmet use rates increased significantly from the 1st to the 2nd Survey. The mean follow-up person-time was 5 months. The IRs for head, face (all portions), and face (upper/mid portions) injuries were higher in non-helmeted than helmeted riders. By using IRs, PAFs for the 3 injuries among the riders in both groups of helmet owners were 77%, 22%, and 32% respectively. The PAFs for each of the above injuries decreased by about 10% as helmet use rates increased. The magnitude of and changes in preventable head and facial injuries following free bicycle helmet distribution and education among helmeted riders was elucidated in this Head Start preschool children population.

Real-Time Safety Helmet Detection Using Yolov5 at Construction Sites

The construction industry has always remained the economic and social backbone of any country in the world where occupational health and safety(OHS)is of prime importance.Like in other developing countries,this industry pays very little,rather negligible attention to OHS practices in Pakistan,resulting in the occurrence of a wide variety of accidents,mishaps,and near-misses every year.One of the major causes of such mishaps is the non-wearing of safety helmets(hard hats)at construction sites where falling objects from a height are unavoid-able.In most cases,this leads to serious brain injuries in people present at the site in general and the workers in particular.It is one of the leading causes of human fatalities at construction sites.In the United States,the Occupational Safety and Health Administration(OSHA)requires construction companies through safety laws to ensure the use of well-defined personal protective equipment(PPE).It has long been a problem to ensure the use of PPE because round-the-clock human monitoring is not possible.However,such monitoring through technological aids or automated tools is very much possible.The present study describes a systema-tic strategy based on deep learning(DL)models built on the You-Only-Look-Once(YOLOV5)architecture that could be used for monitoring workers’hard hats in real-time.It can indicate whether a worker is wearing a hat or not.The proposed system usesfive different models of the YOLOV5,namely YOLOV5n,YOLOv5s,YOLOv5 m,YOLOv5l,and YOLOv5x for object detection with the support of PyTorch,involving 7063 images.The results of the study show that among the DL models,the YOLOV5x has a high performance of 95.8%in terms of the mAP,while the YOLOV5n has the fastest detection speed of 70.4 frames per second(FPS).The proposed model can be successfully used in practice to recognize the hard hat worn by a worker.

An Examination of Concussion Injury Rates in Various Models of Football Helmets in NCAA Football Athletes

While newer, advanced helmet models have been designed with the intentions of decreasing concussions, very little research exists on injury rates in various football helmets at the collegiate level. The aim of this study was to examine concussion injury rates in various models of football helmets in collegiate football athletes. In addition, to compare injury rates of newer, advanced football helmets to older, traditional helmets among collegiate football athletes, a total of 209 concussions and 563,701 AEs （athlete-exposures） among 2,107 collegiate football athletes in seven helmet models were included in the analyses. Concussion injury rates revealed that the Riddell Revolution~ had the highest rate of 0.41 concussions per 1,000 AEs. The Schutt ION 4DTM helmet had the lowest rate of 0.25 concussions per 1,000 AEs. These newer helmet models did not significantly differ from one another （P = 0.74）, however, all models significantly differed from the older, traditional helmet model （P 〈 0.001）. The findings of this study suggest that concussion rates do not differ between newer and more advanced helmet models. More importantly, there are currently no helmets available to prevent concussions from occurring in football athletes.

A method for detecting miners based on helmets detection in underground coal mine videos

In order to monitor dangerous areas in coal mines automatically,we propose to detect helmets from underground coal mine videos for detecting miners.This method can overcome the impact of similarity between the targets and their background.We constructed standard images of helmets,extracted four directional features,modeled the distribution of these features using a Gaussian function and separated local images of frames into helmet and non-helmet classes.Out experimental results show that this method can detect helmets effectively.The detection rate was 83.7%.

基于改进YOLOv5的安全帽检测算法

为了解决建筑工地、隧道、煤矿等施工场景中现有安全帽检测算法对于小目标、密集目标以及复杂环境下的检测精度低的问题,设计实现了一种基于YOLOv5的改进目标检测算法,记为YOLOv5-GBCW.首先使用Ghost卷积对骨干网络进行重构,使得模型的复杂度有了显著降低;其次使用双向特征金字塔网络(BiFPN)加强特征融合,使得算法对小目标准确率提升;引入坐标注意力(Coordinate attention)模块,能够将注意力资源分配给关键区域,从而在复杂环境中降低背景的干扰;最后提出了Beta-WIoU作为边框损失函数,采用动态非单调聚焦机制并引入对锚框特征的计算,提升预测框的准确率,同时加速模型收敛.为了验证算法的可行性,以课题组收集的安全帽数据集为基础,选用了多种经典算法进行对比,并且进行了消融实验,探究各个改进模块的提升效果.实验结果表明:改进算法YOLOv5-GBCW相较于YOLOv5s算法,算法平均精确率(IOU=0.5)提升了5.8%,达到了94.5%,检测速度达到了124.6 FPS(每秒处理帧数),模型更加轻量化,在复杂环境、密集场景和小目标场景下检测能力提升显著,并且同时满足安全帽检测精度和实时性的要求,给复杂施工环境下安全帽检测提供了一种新的方法.

改进YOLOv8s与DeepSORT的矿工帽带检测及人员跟踪

不系帽带,安全帽等于没戴。然而现有的安全帽检测方法,缺乏对帽带异常佩戴的检测研究。针对此问题,结合煤矿井下特殊的作业环境,以人员安全帽帽带检测及人员跟踪为研究对象,提出了CM-YOLOv8s算法检测安全帽及其帽带,利用DeepSORT算法对未系帽带的作业人员进行跟踪。利用井下监控视频制作数据集,使用CM-YOLOv8s对井下人员安全帽帽带进行检测:在YOLOv8s的基础上引入更高分辨率的特征图并新增了一种级联查询机制,在不提高计算成本的前提下能完成对小物体更精准的检测。利用改进DeepSORT对人员进行编码追踪:采用更深层卷积替换DeepSORT中小型残差网络来强化外观信息提取能力。通过自制井下安全帽帽带检测及跟踪数据集对改进算法进行验证,实验结果表明:CM-YOLOv8s的安全帽帽带识别算法平均精度均值达到92.3%,较YOLOv8s提高4.2个百分点。此外,基于CM-YOLOv8s与DeepSORT的安全帽规范佩戴识别系统的平均准确率为85.37%,检测速度达到59 FPS。提出的安全帽帽带检测算法,通过检测帽带是否在人员下颚附近来鉴别安全帽是否规范佩戴,能较好地平衡检测速度与精度,并能适应复杂的井下环境。通过在陈四楼煤矿数月的应用表明,实现了对安全帽佩戴异常的监测预警,加强了对矿工规范佩戴安全帽的有效监管。

基于舱外航天服头盔面窗透明件的边缘复合连接技术研究

舱外航天服头盔为多层面窗结构,压力防护是其首要的防护功能,面窗与金属盔壳之间的连接方式是关键。基于舱外服头盔面窗为聚碳酸酯基材、非规则曲面、端口弧线闭合、薄壁透明等结构特点,提出了一种聚碳酸酯面窗透明件的边缘复合连接方式,并开展了结构设计、仿真分析、原理样机研制、试验验证等工作,研究结果表明:该复合连接结构稳定可靠,性能满足应用要求。

基于Raspberry Pi的安全帽识别系统设计与实现

为了便于施工危险区域人员的自动化识别,提出了一种基于Raspberry Pi的安全帽识别算法。该算法将摄像头采集到的原始视频图像进行滤波、形态学等处理,再对图像中的安全帽进行识别。对于彩色安全帽,将原始图像转换到HSV空间,根据不同颜色色调阈值的设定同时识别红、黄、蓝三种颜色的安全帽,并结合形状特征剔除错误目标。对于白色安全帽,将原始图像转化成B通道下的灰度图像,解决了将黄色误检为白色的问题。采用V通道直方图均衡化的方法,提升了昏暗光线条件下的图像亮度。实验结果表明:在无需提前训练的情况下,算法对于单色安全帽识别准确率超过了91%,对于多色安全帽识别率超过了90%,为施工危险区域的安全隐患排查和作业管控提供了解决方案。

基于上下文融合和注意力的安全帽检测方法

安全帽检测是近年来目标检测在工业生产作业领域的一个研究热点。针对安全帽检测过程中容易出现的小尺度目标错检、漏检等问题,提出一种基于上下文融合和注意力的安全帽检测方法。方法通过利用混合域注意力强调目标关键特征信息,加强特征提取;同时,构建基于非局部注意模块的上下文信息融合结构,将底层全局上下文信息引入深层特征中,进一步细化深层语义信息;此外,利用感受野模块捕获多尺度特征和增大感受野,以减少小尺度目标在特征融合过程中出现特征信息丢失,以及预测过程中对小尺度目标不敏感的问题。实验分析表明,上述方法在安全帽佩戴数据集上对于安全帽检测的AP值达到93.10%,较原YOLOv4提升2.12%,mAP达到93.07%,较原YOLOv4提升1.39%。

复杂作业场景下的反光衣和安全帽检测方法

针对现有算法在复杂的工地环境中进行反光衣和安全帽检测时存在的无法有效区分目标和背景的微小差异问题,提出了一种改进YOLOX的反光衣和安全帽检测算法。首先,将主干网络中空间金字塔池化中的最大池化替换为平均池化,减少特征图的信息损失和过拟合风险;其次,设计一种带权注意力模块(Weighted Convolutional Block Attention Module,W-CBAM)嵌入特征融合层,通过权重系数提升对特征图空间维度的关注,增强特征图的表达能力;最后,添加自适应特征融合(Adaptively Spatial Feature Fusion,ASFF)模块,解决多尺度特征融合时存在的不一致性问题。在扩充后的公开反光衣安全帽数据集的试验结果表明,所提算法精度高达98.79%,优于原始的YOLOX算法和其他先进算法,同时具有较快的检测速度,满足施工环境检测需求。

基于无线传感器网络技术的智能物联施工安全防护系统设计

为解决传统公路施工现场安全防护系统不能实现远程监控的问题,提出了一种基于无线传感器网络技术的智能物联施工安全防护系统,阐述了该安全防护系统的总体结构及工作原理。该系统以安全帽为智能穿戴载体,采用基于ZigBee技术的无线自组网方式,实现了施工现场的实时环境数据采集,通过无线传感器网络技术将环境数据采集至物联网平台进行分析和处理,完成对施工现场的安全监控。实验结果表明,该系统可实现对施工现场安全情况的实时监控,其传输数据稳定可靠,且具备较强的抗干扰能力。

基于多重机制优化YOLOv8的复杂环境下安全帽检测方法

为了解决建筑工地、隧道、煤矿等施工场景中现有安全帽检测算法对于小目标、密集目标以及复杂环境下的检测精度低的问题,提出了一种基于多重机制的安全帽检测方法。以YOLOv8n为基础将Backbone部分的C2f模块加入可扩张残差(DWR)注意力模块,使得网络能够更灵活地适应不同尺度的特征,以而更准确地识别图像中的物体;采用可变形卷积AKConv模块取代主干部分中的原始Conv,为卷积神经网络带来了显著的性能提升,从而实现更高效的特征提取。此外引用了大型可分离核注意力LSKA模块与SPPF结构相结合,大大增强了模型核心的融合能力。在Safety helmet数据集的实验结果表明,改进后的算法相较于原模型,mAP@0.5指标上提升了10.5个百分点,在mAP@0.5-0.95指标上提升了3.7个百分点,能有效提高复杂场景下的安全帽佩戴检测精度。

考虑城乡异质性的电动自行车头盔佩戴行为建模与分析

电动自行车在成为居民短途出行主要交通工具的同时,与其相关的交通安全问题日益突出。正确佩戴头盔可以降低骑行者交通事故的死亡风险,然而统计数据显示,目前全国仍有近一半骑行者在骑行时不佩戴头盔,电动自行车的事故后果无法得到有效控制。较城市而言,乡镇拥有更多的电动自行车保有量,由于道路崎岖、交通建设不完善等原因更容易发生道路交通事故,尤其大量骑行者不佩戴头盔。为明确城市与乡镇骑手头盔佩戴行为异质机理,深入剖析骑手不佩戴头盔的深层原因,运用差异性检验和二项Logistic回归模型,分析骑行者特征、个体头盔认知、执法环境等因素对城市与乡镇骑手头盔佩戴行为的综合影响。研究结果表明:相较于长期生活在城市的群体,乡镇群体的主观安全意识偏弱,对头盔使用时存在的障碍(阻碍视野、不便利性等)较为关注;城市群体对交警执法活动较为敏感,并且在城市中出行时间越长的人越可能佩戴头盔;对于乡镇群体,头盔的保护效能以及价格是他们决策头盔佩戴行为考虑的重要因素。研究结果有助于交警进一步开展安全教育宣传以及执法活动,促进骑行者佩戴头盔安全出行。

改进YOLOX的夜间安全帽检测算法

安全帽检测是保障建筑施工现场安全的一个有效手段。为保证暗光条件下图像分辨度,塔机吊钩摄像头夜间经常需采集灰度图像。由于摄像头晃动和人员走动,安全帽目标区域还经常会出现模糊现象。为解决模糊灰度图像中目标特征丢失所导致的检测精度下降问题,以YOLOX为基准模型,提出一种用于夜间安全帽检测的特征增强和回归权重自适应YOLOX(feature enhancement and regression weight adaptive,FERWA-YOLOX)算法。算法在输入层增加了融合不同大小感受野的多尺度残差(multi-scale residuals,MSR)模块,在同层网络中融合更多局部特征,降低目标局部模糊带来的影响;在解耦头的分类分支增加并行池化通道注意力(parallel pooling channel attention,PPCA)模块,弥补因目标颜色特征丢失所导致的网络分类能力的下降;设计了一种带双惩罚项的损失函数(double penalty items-Siou,DPI-Siou),自适应地降低形状固定目标的形状损失和模糊目标在回归时的权重,提高网络的检测精度。实验结果表明,FERWA-YOLOX较原YOLOX算法,mAP提升了4.88个百分点,参数量仅提升0.5 MB,且满足夜间实时检测需求。