大功率盘形激光焊接飞溅动态识别和特征分析（英文）

提出了一种识别飞溅动态并基于飞溅特征分析焊接状态的方法。以304不锈钢板为试验对象,进行大功率盘形激光平板堆焊试验。利用高速相机捕捉紫外波段和可见光波段的飞溅图像,通过图像处理提取飞溅特征参数,包括质心位置、面积、灰度、平均灰度和半径。基于飞溅特征参数建立飞溅搜索信息库和相似度函数,用于识别飞溅、计算飞溅体积和灰度以及评估焊接状态。通过焊缝宽度与飞溅特征信息的比较,研究焊接状态与飞溅特征参数之间的内在关系。对飞溅特征与焊接状态之间关系进行了试验分析,结果表明,激光焊接过程中飞溅的体积和灰度值增大,焊缝宽度会相应减小,通过飞溅特征参数可有效监测和评估大功率盘形激光焊接过程的状态。     

大功率盘形激光焊飞溅特征分析

焊接过程的在线监控是保证激光焊接质量的关键, 为此, 首先要找到焊接过程传感特征量变化规律以及与焊缝质量间的关系. 飞溅是大功率盘型激光焊接过程中的一个重要现象, 其特征与焊接质量、焊接过程稳定性以及能量利用率等有着密切的联系. 以大功率盘形激光焊接304不锈钢为试验对象, 研究焊接过程中的飞溅特征. 在紫外波段和可见光波段应用高速摄像机摄取焊接过程中产生飞溅的瞬态特征, 通过计算机图像处理技术分析飞溅的数量、面积、行程和质心高度特征参量. 以焊件熔宽作为衡量焊接质量与焊接过程稳定性的因素, 对飞溅特征量进行线性和高次拟合, 研究飞溅特征参量的波动规律, 并与焊件熔宽的变化对比, 探索焊接过程的飞溅特征参量变化规律. 试验结果表明, 根据飞溅特征量变化规律能够对大功率盘形激光焊接304不锈钢板焊接质量做出动态评估, 为实现焊接质量的在线监控提供了试验依据.     

基于LCTF成像选取萝卜叶片的特征波段

从大量的成像光谱数据中选择出有效的特征波段,用于对园艺作物的特征进行快速识别和分类。基于LCTF(liquid crystal tunable filter)成像的优点在于能够连续改变光谱透过中心波长。首先实验选取园艺作物———萝卜叶片为研究对象,获取萝卜叶片表面每隔5nm各波段图像的灰度信息;然后求出各波段的灰度值标准差和相关系数;最后通过波段指数排序选取萝卜叶片的有效特征波段。实验结果表明,在530nm、550nm、535nm、555nm和715nm波段具有较理想的波段指数值,这些波段离散度大,信息量丰富且波段间相关系数小,因此这些波段是识别萝卜叶片的有效特征波段。     

油菜光谱的多重分形分析及叶绿素诊断建模

作物信息科学的重要内容是如何利用作物的信息对其进行无损营养诊断,光谱分析是一种有效可行的途径。对于油菜而言,冠层光谱的特征是描述其营养状况的重要指标。但由于原始光谱总是受到一些如环境、气候等外在因素的影响,其巨大的波动导致难以直接用于油菜生物量的诊断。然而,光谱的多重分形特征将保持相对稳定。为研究油菜冠层光谱与叶绿素含量的关系,基于多重分形理论,提出了基于油菜冠层光谱特征的叶绿素定量预测模型和定性识别模型。以24个移栽种植小区和24个直播种植小区的高油酸油菜苗期样本为试验对象。首先,利用流行的多重分形去趋势波动分析提取了6个不同波段范围内光谱的广义Hurst指数和质量指数及其他相关的特征参数,发现它们都呈现典型的多重分形特性。但两种不同种植方式下的光谱特征也存在差异。接着,通过多重分形特征参数与SPAD值的相关分析发现不同波段的光谱所含的有效信息不同。以多重分形特征参数建立单变量油菜叶片SPAD值预测模型,移栽方式、直播方式及混合样本的预测模型相对均方根误差均小于5%。最后,以多重分形特征组合建立识别模型,以Fisher线性判别法识别移栽和直播两种种植方式的最大约登指数为0.902 5,对应最敏感波段为350~1 350 nm。这项有意义的工作为预测油菜叶绿素提供了理论基础和实践方法,也为寻找敏感波段进行识别诊断提供了有效的途径。     

基于离焦成像的粒子轨迹测速

提出将离焦成像与粒子轨迹测速相结合,分别针对直线型和曲线型粒子轨迹,构建了离焦特征参数σ的识别算法,形成了用于三维速度测量的离焦粒子轨迹测速(DPSV)方法.该方法基于粒子图像离焦特征参数σ沿运动轨迹线的线性变化假设,通过拟合粒子图像的灰度分布来识别σ,从而获得粒子深度信息.在直线轨迹拟合的基础上,进一步采用圆弧拟合处理湍流实验中常见的弯曲轨迹图像.形成了相应的图像处理流程,并采用仿真图像验证了参数识别的准确性,在本文实验装置的噪声水平下相对误差为9.4％左右.采用LED光源和5μm孔径光阑组成的发光点进行模拟实验,验证了深度位置z与σ的线性关系.最后,将DPSV技术应用于射流流场,得到了三维速度分布.     

基于改进U-Net的量子点STM图像分割

为提升量子点图像分割精度,降低特征识别误差,提出一种基于改进U-Net的量子点图像分割方法.首先,在预处理阶段,设计了以色彩通道为权值的灰度化算法,以提升后续分割效果.其次,在STM图像分割部分,在原始U-Net结构上引入中间过渡层以均衡网络各层特征.而后,建立数据集,并通过实验对比不同分割算法的精确度、召回率、F-measure.最后,将分割算法应用于量子点的特征识别,并测试了不同分割方式对应用的影响.实验结果显示,改进灰度化方法保留细节信息丰富,明显提升了量子点分割精度;改进U-Net的平均精确率、召回率、F-measure相较原始网络分别提升了13.83%、2.16%、8.13%.同时,实验数据表明由于分割精度的提升,量子点数量、纵横比等特征参数的识别更加精确.     

高光谱成像与图像结合进行油菜角果蚜虫侵染的定位识别

油菜蚜虫可造成油菜籽的严重减产,及早进行油菜蚜虫判别以及其侵染定位识别有助于精准喷药。采用可见-近红外高光谱成像技术结合图像分析对185个蚜虫侵染以及138个健康油菜角果进行判别,并进行蚜虫的定位分析。首先采用主成分分析法(PCA)对两类样本的平均光谱进行聚类分析,并基于X-loading得出737nm波段可作为判断蚜虫的重要波段,采用Boxplot进行两类样本间单波段处的统计分析,同时得出基于737nm波段判断蚜虫侵染油菜角果的线性公式为y=2.917 6-3.345 7x(x为样本在737nm处的光谱值,y为样本的分类预测值)。采用此公式对实验样本进行判别分析,可以发现角果蚜虫识别率为99.0%。同时基于737nm处的油菜角果单波段灰度图进行蚜虫的定位识别,可以得到蚜虫的识别率为81.1%。结果表明,采用737nm处的单波段光谱信息以及图像信息可进行油菜角果蚜虫侵染的定位识别,为进一步开发便携仪检测仪以及精准喷药提供理论和方法依据。     

基于路径跟踪的水下焊接V形焊缝识别中强干扰的去除

介绍机器人水下焊接V形焊缝图像处理中基于模糊推理的阈值自动选取的算法,提出基于路径跟踪的强干扰的去除方法.制作9种不同形状的窗口模板,以列扫描方式求取这些窗口所覆盖的区域内最大的平均灰度值,同时获取取得最大灰度平均值时的垂直方向的位置,然后根据获取的最大平均灰度值以及相邻区域最大平均灰度值的改变量进行模糊推理,判断获得这些窗口所覆盖区域的阈值.最后跟踪取得最大灰度平均值时的所有位置,在最低位置以下以及最高位置以上检测经阈值处理后的白点,可以消除因水泡、较强弧光形成的干扰.     

基于典型目标反射率的近红外场景仿真

利用积分球对近红外增强相机进行了可见、近红外波段的辐射定标,并采用该相机采集了典型目标的可见、近红外图像.利用便携式地物光谱仪采集了相同目标可见、近红外波段的光谱反射率曲线,并分析了目标在可见、近红外波段的光谱反射特性.通过寻找同一天气条件、同一时间段所拍摄的典型目标近红外图像灰度值与可见光图像灰度值及典型目标可见、近红外波段反射率之间的关系,提出一种基于可见光图像及目标反射率反演近红外图像的算法.在近红外图像反演过程中考虑了大气透过率的影响,反演结果表明,本文算法可以较好地体现典型目标近红外图像的全局特征,为后续不同天气条件下近红外场景仿真的实现提供参考.     

多波段红外图像的海面舰船目标检测

现有的基于单个红外宽波段的海面舰船目标探测系统在面对复杂海天背景、岛岸背景、恶劣天气、亮带干扰或诱饵弹干扰等情况时,系统的探测率、虚警率、探测距离等性能指标均会受到严重的影响;为此,开展了基于多波段红外图像的海面舰船目标检测方法的研究。通过中波红外多波段数据采集系统实际采集107组五个中波红外波段的图像;波段1-5分别为3.7～4.8,3.7～4.1,4.4～4.8,3.7～3.9和4.65～4.75 μm;对多波段图像进行手动标注构建样本数据集,其中,正样本舰船目标298个,负样本非舰船目标353个。对于多波段红外图像,首先进行PCA降维并采用选择性搜索算法生成初始目标候选区域;针对候选区域中存在大量明显的非舰船目标区域的问题,利用积分图像计算候选区域的局部对比度,依据红外舰船目标的几何和灰度特征从初始目标候选区域中筛选出舰船目标可能性大的区域作为舰船目标候选区域。然后对舰船目标候选区域进行拓展以融入局部上下文信息,对于候选区域对应的5波段红外图像,分别提取每个波段图像的稠密SIFT特征,并将128维SIFT特征向量降为64维,融入SIFT特征的空间和波段位置分布信息得到新的特征向量,基于高斯混合模型对候选区域的特征向量集合进行编码融合得到舰船目标候选区域的费舍尔向量表示,最后利用线性SVM分类器识别出舰船目标。对多波段图像进行舰船目标候选区域生成实验,所提出的基于红外舰船目标的几何和灰度特征的约束方法可以有效地克服选择性搜索算法的不足,从初始目标候选区域中快速定位出舰船目标候选区域,对25组多波段图像进行实验,舰船目标候选区域生成的整体耗时为0.353 s,定位舰船目标区域耗时0.005 s。对100个正负样本进行目标识别测试,所提出的目标识别算法融合了目标的多波段图像特征信息,通过引入费舍尔向量挖掘了多波段图像梯度统计特征的深层次信息,算法的识别率达到了0.97,显著高于单波段红外图像的目标识别率。对25组多波段图像进行舰船目标检测实验,所提出的舰船目标检测方法能够在海天背景、岛岸背景以及亮带干扰等不同场景下完成海面舰船目标的检测工作,舰船目标定位准确,舰船目标召回率达到了0.95,每组多波段图像的平均检测耗时为1.33 s。研究结果表明,充分考虑海面舰船目标在红外图像中与局部海洋背景的辐射差异以及有效地融合舰船目标在多个红外波段图像中的辐射特征,可以增强舰船目标的可分性,提高舰船目标的识别率以及检测率,为基于多波段红外图像的海面舰船目标检测提供了新的技术支持。     

Analysis of mechanism of plasma and spatter in CO2 laser welding of galvanized steel

In laser welding, quality, reproducibility and formability are required. That is the great problem in the automation of the laser welding process. Therefore, construction of an on-line process monitoring system of high accuracy is requested.The light which is emitted from plasma and spatter in laser welding was detected by photo-diodes. It was found that the light intensity depends on welding speed, laser power and flow rate of assist gas. The relationship between the plasma and spatter and the bead shape, and the mechanism of plasma and spatter were analyzed for on-line laser weld monitoring systems.     

Real time estimation of CO2 laser weld quality for automotive industry

Laser welding is one of the most precise welding processes in joining sheet metals. In laser welding, performing real time evaluation of the welding quality is very important to enhance the efficiency of the welding process. In this study, the plasma and spatter, which are generated during laser welding, are measured using UV and IR photodiodes. The factors that influence weld quality are classified into five categories; optimal heat input, slightly low heat input, low heat input, partial joining due to gap mismatch, and nozzle deviation. The data number deviated from reference signals and their standard deviations were also considered to evaluate the qualities. A system was also formulated to perform real time evaluations of the weld quality using a fuzzy multi-feature pattern recognition with the measured signals.     

盘型激光焊金属蒸汽特征色调-色饱和度-亮度分析法研究

对于大功率盘型激光焊,金属焊件表面在激光束辐射下强烈汽化并形成等离子体状的金属蒸汽羽状物. 该金属蒸汽羽状物可逆向激光束传输,对激光有明显的屏蔽作用, 降低激光辐射至焊件的能量密度,影响焊接效率和质量. 因此研究金属蒸汽特征变化规律及其与焊接质量之间的关联 ,可实现由金属蒸汽特征实时监测激光焊接状态. 以10 kW大功率盘型连续激光焊接304不锈钢钢板为试验对象, 应用高速摄像机摄取金属蒸汽动态图像,将其转换至色调-色饱和度-亮度空间, 提取金属蒸汽面积、激光束受影响路径长度等相关特征量, 以焊缝熔宽的变化作为衡量焊接状态稳定性的参数. 通过金属蒸汽特征值的均值统计和方差分析,试验证明根据金属蒸汽面积和激光束受影响路径长度等金属蒸汽特征可有效地反映熔宽质量, 从而对焊接状况做出动态评估.     

Optimization of hybrid laser – TIG welding of 316LN steel using response surface methodology (RSM)

In the present study, the hybrid laser – TIG welding parameters for welding of 316LN austenitic stainless steel have been investigated by combining a pulsed laser beam with a TIG welding heat source at the weld pool. Laser power, pulse frequency, pulse duration, TIG current were presumed as the welding process parameters whereas weld bead width, weld cross-sectional area and depth of penetration (DOP) were considered as the process responses. Central composite design was used to complete the design matrix and welding experiments were conducted based on the design matrix. Weld bead measurements were then carried out to generate the dataset. Multiple regression models correlating the process parameters with the responses have been developed. The accuracy of the models were found to be good. Then, the desirability approach optimization technique was employed for determining the optimum process parameters to obtain the desired weld bead profile. Validation experiments were then carried out from the determined optimum process parameters. There was good agreement between the predicted and measured values.     

Research on air tightness of image intensifier tube during laser sealing北大核心CSCD

采用高功率YAG激光焊接机对高性能三代像管管壳后端(4J34可伐合金)与荧光屏屏环(4J49可伐合金)进行封接试验。研究了激光功率、脉冲宽度对焊接接头成型及表面热扩散的影响规律。研究表明:4J34合金与4J49合金表面成型质量在设备最大工作电流100 A,激光功率195 W及脉宽1.7 ms时最好,相对于激光功率,脉冲宽度对焊缝熔宽和熔深的影响更加显著,接头焊接中心区硬化最为严重,其硬度最大,热影响区次之。     

Vision-based weld pool boundary extraction and width measurement during keyhole fiber laser welding

In keyhole fiber laser welding processes, the weld pool behavior is essential to determining welding quality. To better observe and control the welding process, the accurate extraction of the weld pool boundary as well as the width is required. This work presents a weld pool edge detection technique based on an off axial green illumination laser and a coaxial image capturing system that consists of a CMOS camera and optic filters. According to the difference of image quality, a complete developed edge detection algorithm is proposed based on the local maximum gradient of greyness searching approach and linear interpolation. The extracted weld pool geometry and the width are validated by the actual welding width measurement and predictions by a numerical multi-phase model.     

Prediction of weld strength and seam width for laser transmission welding of thermoplastic using response surface methodology

In the present work, a study is made to investigate the effects of process parameters, namely, laser power, welding speed, size of the laser beam and clamp pressure, on the lap-shear strength and weld-seam width for laser transmission welding of acrylic (polymethyl methacrylate), using a diode laser system. Response surface methodology (RSM) is employed to develop mathematical relationships between the welding process parameters and the output variables of the weld joint to determine the welding input parameters that lead to the desired weld quality. In addition, using response surface plots, the interaction effects of process parameters on the responses are analyzed and discussed. The statistical software Design-Expert v7 is used to establish the design matrix and to obtain the regression equations. The developed mathematical models are tested by analysis-of-variance (ANOVA) method to check their adequacy. Finally, a comparison is made between measured and calculated results, which are in good agreement. This indicates that the developed models can predict the responses adequately within the limits of welding parameters being used.     

Measurement of micro weld joint position based on magneto-optical imaging

In a laser butt joint welding process,it is required that the laser beam focus should be controlled to follow the weld joint path accurately.Small focus wandering off the weld joint may result in insufficient penetration or unacceptable welds.Recognition of joint position offset,which describes the deviation between the laser beam focus and the weld joint,is important for adjusting the laser beam focus and obtaining high quality welds.A new method based on the magneto-optical(MO)imaging is applied to measure the micro weld joint whose gap is less than 0.2 mm.The weldments are excited by an external magnetic field,and an MO sensor based on principle of Faraday magneto effect is used to capture the weld joint images.A sequence of MO images which are tested under different magnetic field intensities and different weld joint widths are acquired.By analyzing the MO image characteristics and extracting the weld joint features,the influence of magnetic field intensity and weld joint width on the MO images and detection of weld joint position is observed and summarized.     

大功率光纤激光焊焊缝跟踪偏差红外检测方法

精确控制激光束使其始终对准并跟踪焊缝是保证激光焊接质量的前提.针对大功率(激光功率10 kW)光纤激光焊接304型不锈钢紧密对接焊缝(间隙为0-0.1 mm),研究一种基于红外热像的焊缝跟踪偏差检测新方法. 采用红外传感高速摄像机摄取焊接区域熔池红外动态热像,分析激光束对准和偏离焊缝中心时的熔池温度分布和红外辐射特性,以熔池匙孔形变参数和热堆积效应参数作为激光束与焊缝中心偏差检测特征值,通过图像识别技术研究和分析特征值与焊缝偏差之间的关系. 激光焊接试验结果表明,熔池匙孔形变参数和热堆积效应参数与焊缝偏差 关键词： 大功率光纤激光焊 焊缝跟踪偏差 红外热像 检测