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.

Tribological Properties of the Plating Chromic Layer for 38CrMoAl Cylinder Liner

A chromium layer about 100 μm thickness was plated on the 38CrMoAl cylinder liner and the chromium layer was mcro quilting milled by using quilting grinding machine. The tribological properties and wear comparison test were studied. The friction coefficient of the cylinder liner plated chromic layer and micro quilting milled is 15%-30% lower than the ordinary cylinder liner. The pits generated by micro quilting milling on the chromic layer surface had good effect of accommodating the abrasive grains and storaging lubricants, which improved the effect of the friction pair significantly. The single-cylinder machine run-in tests revealed that the cylinder liner with plated chromic layer and micro quilting milling had good wear durability, and was different wear mechanisms to ordinary cylinder liner.

Experimental Study on Wear Performance and Oil Film Characteristics of Surface Textured Cylinder Liner in Marine Diesel Engine

It is of a vital importance to reduce the frictional losses in marine diesel engines. Advanced surface textures have provided an e ective solution to friction performance of rubbing pairs due to the rapid development of surface engineering techniques. However,the mechanisms through which textured patterns and texturing methods prove beneficial remains unclear. To address this issue,the tribological system of the cylinder liner?piston ring(CLPR) is investigated in this work. Two types of surface textures(Micro concave,Micro V?groove) are processed on the cylinder specimen using di erent processing methods. Comparative study on the friction coe cients,worn surface texture features and oil film characteristics are performed. The results demonstrate that the processing method of surface texture a ect the performance of the CLPR pairs under the specific testing conditions. In addition the micro V?groove processed by CNCPM is more favorable for improving the wear performances at the low load,while the micro?con?cave processed by CE is more favorable for improving the wear performances at the high load. These findings are in helping to understand the e ect of surface texture on wear performance of CLPR.

Application of Optical Visualization Method in Observing Flow Field Around a Cylinder

In this paper, a free-surface synthetic schlieren(FS-SS) method was performed to detect the free surface disturbances. It is a purely optical method that uses refraction of light to reconstruct the height changes of water surface. The theory was developed based on Moisy's research, but has mainly been used in small-scale applications like painting and coating industry. Based on the methods and theories of the literature review, an in-depth investigation was conducted to optimize the FS-SS method and verify its feasibility on a relatively large-scale(e.g., wake region of shallow flow). The experimental setup was simplified which is approachable in most laboratories. Through proper experimental setting and an optimized post-processing routine, the quality of image was highly improved and ensured the accuracy of results. A drop test was performed proving the continuity of FS-SS method in the time domain. Also, a comparison test with flow around a cylinder at two speeds showed the ability of FS-SS method to reconstruct the irregular water surface in relative large-scale flow structures.

Syntheses and Tribological Property of CrMoN/MoS2 Multilayer Films on Piston Rings of Heavy Vehicle Engine

In order to prolong the service life of piston rings of heavy vehicle engine and decrease the friction and wear of piston rings and cylinder liner,CrMoN/MoS_2 multilayer films were deposited on the surface of rings by magnetron sputtering and low temperature ion sulfuration.FESEM equipped with EDX was adopted to analyze the compositions and morphologies of surface,cross-section,and wear scars of the multilayer films.The nano-hardness and Young＇s modulus of the films were measured by a nano tester.Tribologicalproperties of the films were tested by an SRV~174;4 wear tester.The experimentalresults indicate that the structures of the multilayer films are dense and compact.The films possess nano hardness value of approximately 26.7 GPa and superior ability of plastic deformation resistance.The multilayer films can activate solid lubricating,and possess an excellent antifriction and wear resistance under the conditions of heavy load,high frequency,high temperature,and dynamic load.

Friction Wearing Properties Research of High-silicon Aluminium Alloy Cylinder Sleeve

With tank special purpose lubricating oil,100N,200N and 800N loading force and different loading times,spray-formed high-silicon aluminium alloy and military superpower engine steel cylinder sleeve materials were used for comparative friction test and friction pair comparision test under simulated engine work condition.The results showed that,compared with steel cylinder sleeve materials,high-silicon aluminium alloy showed more excellent wearing resistance.The friction mechanism analysis of high-silicon aluminium alloy indicated that high-hardness particles in soft parent metal had determinative function,including wearing resistance and supporting ability when wearing happened.Dents on soft parent metal surface produced by friction could store oil and were helpful for lubrication.The friction trace analysis showed that,high-hardness particles in high-silicon aluminium alloy could produce friction trace on 42MnCr52 steel surface,which proved friction function of high-hardness particles in high-silicon aluminium alloy.

Real-time Safety Helmet-wearing Detection Based on Improved YOLOv5

Safety helmet-wearing detection is an essential part of the intelligentmonitoring system. To improve the speed and accuracy of detection, especiallysmall targets and occluded objects, it presents a novel and efficient detectormodel. The underlying core algorithm of this model adopts the YOLOv5 (YouOnly Look Once version 5) network with the best comprehensive detection performance. It is improved by adding an attention mechanism, a CIoU (CompleteIntersection Over Union) Loss function, and the Mish activation function. First,it applies the attention mechanism in the feature extraction. The network can learnthe weight of each channel independently and enhance the information dissemination between features. Second, it adopts CIoU loss function to achieve accuratebounding box regression. Third, it utilizes Mish activation function to improvedetection accuracy and generalization ability. It builds a safety helmet-wearingdetection data set containing more than 10,000 images collected from the Internetfor preprocessing. On the self-made helmet wearing test data set, the averageaccuracy of the helmet detection of the proposed algorithm is 96.7%, which is1.9% higher than that of the YOLOv5 algorithm. It meets the accuracy requirements of the helmet-wearing detection under construction scenarios.

低硫条件下船舶主机缸套-活塞环磨损现状与应对策略

国际海事组织(IMO)的“限硫令”于2020年1月在全球范围内实施后,燃用低硫燃油(LSFO)成为许多远洋公司的主要选择。对多家远洋公司船舶主机燃用LSFO后的运行状况进行调研,发现缸套、活塞环等零部件存在异常磨损及故障显著增多的现象。将LSFO对船舶主机的影响以及其实际应用案例进行分析与总结,并针对燃用LSFO的主机所出现的异常磨损和故障现象提出应对策略。该研究有助于解决LSFO条件下船舶主机缸套和活塞环异常磨损问题的实际需求,希望进一步推动LSFO条件下缸套和活塞环异常磨损的机理研究。

真空扩散对奥氏体灰铸铁渗硼层组织与性能影响的研究

为了改善奥氏体灰铸铁渗硼层厚度不足和脆性过大的问题,研究了渗硼处理和渗硼/真空扩散复合处理对奥氏体灰铸铁表面改性层摩擦磨损性能的影响和影响机理。利用光学显微镜、能谱仪、X射线衍射仪、电子探针等设备对单一处理和复合处理试样的渗层进行了金相组织、元素分布、物相组成、表面硬度、截面硬度以及磨痕分析。结果表明:真空扩散对渗硼层有着增厚的作用,能够有效的将脆性较大的FeB相转化为脆性相对较小Fe2B相,改善渗硼层的耐磨性,但会使得渗层的表面硬度略微降低,渗层中石墨的粗化现象更为加重。

基于有限元分析的液压缸密封圈磨损分析

液压缸密封圈磨损原因较多,难以通过肉眼直观观察,很难找到直接原因,提出基于有限元的液压缸密封圈磨损量分析方法。基于液压缸密封圈几何模型、材料模型和物理模型,分析了橡胶密封圈在应力和应变上的非线性,通过定义应变密度函数,构建了液压缸密封圈的有限元模型。根据液压缸活塞的密封结构,分析了液压缸活塞的运动特点,利用Archard模型描述密封圈的外观特征变化,通过计算密封圈接触面上任意一点的磨损深度,分析液压缸密封圈的磨损量。试验结果表明,所研究方法可以对液压缸密封圈的裂纹深度与裂纹扩展角度进行有效分析,两项指标的试验分析与实测结果基本一致,裂纹深度的试验分析误差最大值仅为0.002 mm。方法有助于提高液压缸设计的密封可靠性。

基于改进阈值分割算法的刀具损伤检测方法

为了解决刀具损伤缺陷难以被视觉检测系统收集的问题,提出了一种基于粒子群算法(PSO)的Otsu阈值分割法对刀具磨损量进行检测。算法改进了PSO算法惯性系数的更新策略,有效扩大了算法的搜索范围,缩短了算法的运行时间,通过对粒子群添加扰动,解决了传统粒子群算法容易陷入局部最优的问题,搭建了实验平台,验证所用检测方法的有效性。实验结果表明,该检验方法能够实现刀具损伤区域识别和刀具损伤量的测量,而且相较于Otsu、Canny算法,局部阈值分割法等算法具有识别精度高,运行速度快等优点。研究结果对于实际刀具缺陷检测系统具有一定的参考价值。

结构强度试验液压缸振动智能检测及定位方法

针对结构强度试验加载过程中液压缸振动问题,提出了一种液压缸振动智能检测及定位方法,通过对常见液压缸振动信号的捕捉建立异常声音信号专家知识库,同时在试验现场布置声音检测传感器阵列以获取液压缸振动信号,通过与专家知识库匹配确定异响来源并分析可能的原因。引入液压缸空间分布坐标,对可能发生振动的液压缸进行快速定位,对液压缸振动的位置和原因进行快速检测判断,起到辅助试验实施、降低试验风险的作用。

基于改进凸包缺陷算法的扎带定位方法

针对传统视觉在检测捆扎带时存在难检、漏检、定位扎带位置困难和速度慢等问题,提出一种基于改进凸包缺陷算法的扎带定位方法。采用自适应直方图均衡化图像增强算法,以增强编织袋区域与背景的对比度;利用快速凸包算法获得编织袋轮廓的凸包,减少获取凸包的时间;最后,通过改进凸包缺陷算法对编织袋轮廓凸包进行缺陷检测,根据检测结果中凸包缺陷点的位置和缺陷深度进行筛选得到扎带定位点。同时为验证该算法的准确性与鲁棒性,在具有复杂背景干扰的环境下进行实验,将传统凸包缺陷算法与改进后的凸包缺陷算法进行对比分析。实验结果表明:相比于传统凸包缺陷算法无法检测出编织袋轮廓的全部缺陷从而存在漏检,改进后的凸包缺陷算法漏检率为0,定位误差小于4 mm,可有效定位扎带位置并具有较高的鲁棒性。

合金灰口铸铁气缸套组织形貌与耐磨性研究

针对5种成分和组织的灰口铸铁气缸套,包括2种贝氏体铸铁气缸套和3种珠光体铸铁气缸套,全面分析了其成分、相组成及硬度等材料特性,将喷钼活塞环与5种合金灰口铸铁气缸套对磨,通过比较各配对副的摩擦磨损性能,获得了喷钼活塞环与5种合金灰口铸铁气缸套的匹配规律。研究结果表明:5种气缸套材料均为由硬质相、基体和石墨构成的复合耐磨结构,各组分的性能、分布及其配比,是决定耐磨性的关键因素。珠光体基体气缸套的磨损量相对稳定,气缸套与活塞环的相对磨损量正常;铸态贝氏体气缸套磨损较轻,配对活塞环的磨损量比珠光体气缸套大;贝氏体气缸套和活塞环的磨损均量很小。

基于YOLOv7-tiny改进的口罩佩戴检测算法YOLOv7-DSC

针对密集人群下口罩佩戴检测实时性差、难以部署到移动端的问题,提出基于YOLOv7-tiny改进的口罩佩戴检测算法YOLOv7-DSC。该算法结合深度可分离卷积改进的SE注意力机制设计了一种轻量化特征提取模块,并结合BiFPN设计了一种加权特征融合模块。经实验验证,YOLOv7-DSC算法在口罩数据集上mAP为96.9%,与YOLOv7-tiny算法相比仅降低了0.5%;相比于YOLOv3-tiny、YOLOv4-tiny、YOLOv5s、MobileNetV3、ShuffleNetV2、GhostNet和Swin-Transformer算法在mAP上分别高出13.4%、11.2%、4.5%、5.7%、5.8%、4.2%和5.1%;在检测精度与YOLOv7-tiny算法相当的情况下,参数量和计算量分别减少了60%和55%,仅为2.4 M和6.0 G,极大地降低了硬件成本。

细长竹木圆棒双激光对射自动检测系统设计

针对细长竹木圆棒胚体存在的表面凹坑、开裂等外观缺陷,设计了一种双激光对射式自动检测系统。机械结构采用载物链条使细长竹木圆棒整体平移并在弹性压板下滚动,由对称布置的双激光斜入射至压板和圆棒贴合部位,实现对竹木圆棒表面的双螺旋扫描和空隙检测;控制部分以西门子S7-200 Smart PLC为主体,设计了与实体链条同步的数字链条进行并行控制。测试结果表明,系统相对于人工检测漏检率为0,具有成本低、效率高、可靠性高、易于使用等特点,可有效解决细长竹木圆棒胚体表面质量初筛问题。

FLM-YOLOv8:一种轻量级的口罩佩戴检测算法

针对现有的口罩佩戴检测模型无法较好平衡检测精度与速度,参数量较大,漏检和误检率高等问题,提出了一种轻量级的口罩佩戴检测算法FLM-YOLOv8。使用轻量级FasterNet替换YOLOv8n的主干特征提取网络,提升网络检测速度;融合FasterNet Block改进C2f模块,降低模型计算复杂度;提出SPPF-LSKA结构,增强模型的特征表达能力和感知能力,提高网络检测精度;设计Inner-MPDIoU边界框回归损失函数,提高回归预测精度,加快收敛速度。创建标注了一个复杂多元场景下的口罩佩戴数据集,并使用马赛克数据增强,以提高网络泛化能力。实验结果表明,该算法在正确佩戴口罩、未正确佩戴口罩和未佩戴口罩目标上的mAP@0.5达到了91.3%,FPS达到了143.6,实现了更加实时准确的口罩佩戴检测。

基于声发射的缸套珩磨网纹磨损深度检测

珩磨网纹对维持活塞-缸套间润滑状态至关重要,磨损造成的网纹深度变浅会引起润滑恶化并加剧摩擦磨损,甚至造成拉缸等重大机械事故。提出一种基于声发射的缸套珩磨网纹磨损深度检测方法,从声发射信号中识别提取了网纹深度有关特征变量,通过构建卷积神经网络模型建立了声发射特征信号与网纹深度的映射关系。基于长行程往复式摩擦试验机,对不同网纹简约谷深的缸套试样在变转速、变载荷工况下实测的声发射信号进行采集分析,发现网纹简约谷深与声发射信号低频和高频段存在较强相关性。基于所建模型对早期缸套网纹磨损进行了检测评估,结果表明:珩磨网纹深度识别平均准确率达94%,证实了基于声发射技术进行缸套网纹磨损深度非介入式检测的可行性。

某新型发动机气缸磨损分析及优化设计

某新型高强化发动机在设计开发过程中出现曲轴箱压力异常升高、窜气量大的问题。经故障再现和分析,确认故障原因为气缸局部异常磨损。通过成分检测、性能测试、热机耦合CAE对比分析,确定根本原因为气缸在高的热负荷、恶劣的润滑条件下承受复杂作用力而出现“抛光”现象。基于根本原因提出了优化设计方案,耐久试验结果表明,优化后的气缸状态良好,再未出现气缸“抛光”的问题。

基于小波分析的智能轮胎磨损和载荷的检测方法

为了获取轮胎运行状态下的状态信息,提出了一种基于小波变换的智能轮胎磨损和载荷检测方法,用于实时检测轮胎的载荷和磨损状态。为研究轮胎在运行时胎面的震动情况,通过贴装于轮胎内壁中央的三轴加速度传感器获取加速度波形,再通过莫尔斯小波多尺度分析轮胎接地点处的胎面震动情况,提取特征,输入BP神经网络,获取检测结果。结果表明:所提出的算法可以较为精确地监测轮胎的磨损和载荷状态;在90%以上的情况下,可以实现磨损绝对误差在0.3 mm以内,载荷绝对误差在12 kg以内。为车辆提供关键轮胎信息,助力安全驾驶,有广泛的应用前景。