焊接视觉图像质心算法及其特性

从焊接视觉图像质心计算公式出发,根据图像灰度与视觉传感器响应电压值的关系、响应电压值与检测部位的发射率及光谱辐射强度等关系,由熔池特性,推导并建立了焊接视觉图像质心与焊接区各点温度间关系模型.通过焊接温度场的数值化,得到图像质心与焊缝中心关系模型的数值解及关系曲线,证明了视觉图像质心与焊缝中心存在相关性,为基于图像质心的焊缝跟踪算法提供理论依据.分析了质心算法的特性,为进一步研究实时焊接图像的质心及算法打下理论基础.     

BP神经网络在焊缝位置识别中的应用

研究了一种焊缝位置识别新方法,在一定工艺条件下,使用视觉传感器采集焊接熔池图像,选取图像中熔池前端部分进行处理,先对其进行中值滤波与灰度变换,在此基础上,获取每一幅熔池图像的质心值、质心位移、质心速度及电弧与焊缝的偏差值作为训练样本数据.以质心值、质心位移和质心速度为输入量,以偏差值为输出量,利用BP神经网络建立其数学模型,最后对该模型进行检验.检验结果表明,该模型能够较准确地描述熔池图像质心与焊缝偏差之间的关系,为进一步实现精确的焊缝跟踪提供了理论和试验依据.     

一种基于图像质心的焊缝跟踪新方法

研究一种基于图像质心识别的电弧焊焊缝跟踪新方法。通过视觉传感器获取焊接区熔池图像，抽取图像质心坐标并构成状态向量，建立一种基于图像质心的状态方程和位置测量方程。在此基础上，应用卡尔曼滤波对图像质心位置进行状态估计，在时域中采取递推计算的方式得到最小均方差条件下的焊缝位置最佳预测值．从而消除过程噪声和测量噪声引起的焊缝位置测量偏差。计算机仿真和实际焊接试验结果显示该方法可有效地提高焊缝跟踪精度。     

BP神经网络修正熔池红外图像焊缝宽度测量算法

在激光焊接中,焊缝宽度的实时动态变化对于描述焊接质量起着至关重要的作用。焊缝宽度的准确测量有助于理解焊接过程,获得焊接质量控制模型。针对大功率光纤激光焊接304型不锈钢过程,利用高速摄像机,获得清晰的熔池动态红外图像。红外图像仅仅是熔融焊缝处的热成像,难以准确测量焊缝宽度,必须利用BP神经网络加以修正,得到实际焊缝宽度。三组试验结果表明了BP神经网络焊缝宽度测量模型的有效性。     

焊缝位置识别的图像处理方法设计

在基于视觉的自动焊接中，由于焊缝图像是复杂多变的，要准确得到焊缝位置选择合理高效的处理算法是非常重要的。针对这一问题为焊缝图像提出了一系列处理步骤，它包括中值滤波、自适应阈值二值化、孤点滤波、边缘检测，焊缝位置搜索五步，其中详细叙述了重要的自适应阈值二值化与孤点滤波算法。在每一步处理中都用了四幅完全不同的焊缝图像做了观察和对比，结果显示所选算法产生较好的处理效果。     

基于熔池红外图像实时跟踪焊缝宽度算法研究

在激光焊接过程中,激光焊机所附带的传感器能够准确实时地反映焊接过程,焊缝宽度的实时动态变化对于描述焊接质量起着至关重要的作用。焊缝宽度准确测量的研究有助于理解焊接过程,获得焊接质量控制模型。针对大功率光纤激光焊接,利用恰当的滤镜系统获得清晰的焊缝熔池图像,建立熔池红外传感跟踪系统。针对大功率光纤激光焊接304型不锈钢过程,研究一种通过熔池图像实时检测焊缝宽度变化的方法。利用高速摄像机获得熔池动态红外图像,通过一系列图像处理技术得到焊缝宽度的识别模型。解决了在金属处于高温下熔池与周边非熔化区的温度比较接近,很难用红外摄像来准确测量焊缝宽度的难题。试验结果表明,所建立的焊缝实际宽度测量模型测量效果良好。     

电弧焊接焊缝偏差方法研究

焊缝跟踪技术是自动电弧焊接的一个重要研究领域,实现精确的焊缝跟踪对于提高焊接质量具有非常重要的作用。而要实现精确的焊缝跟踪,焊缝偏差(即焊缝中心与电弧的偏差)检测技术是一个关键。通过图像处理技术,选取熔池图像处理区域(包括熔池前端与熔池前端部份焊缝),并将熔池图像质心作为分析焊缝偏差的特性参量,研究利用熔池特性参数来建立焊缝偏差测量视觉模型的方法。     

焊接理论/一般问题

建立了一种基于熔池图像质心的焊缝位置测量模型，通过视觉传感器获取焊接区熔池图像，选择熔池前端为处理区域，对该区域进行中值滤波与图像灰度变换，并计算该区域的熔池图像质心值及相对应的焊缝偏差。在不同的焊接条件下，[第一段]     

基于视觉图像质心的焊缝跟踪新技术

从基于图像质心的焊缝跟踪技术出发，通过对图像质心公式的深入研究，根据图像灰度与视觉传感器响应电压值的关系、响应电压值与检测部位的发射率及光谱辐射强度等关系，从理论上得出质心与焊缝中心的关系模型，论证了图像质心与焊缝中心存在相关性，并通过大量试验证明了关系模型的准确性。     

空间多位置摆动激光填丝焊接熔池动态行为及焊缝成形

以316L奥氏体不锈钢管道为研究对象,在摆动激光焊接研究基础上,对管道多位置激光填丝焊接熔滴过渡和焊缝成形展开研究,分析焊接熔池动态特征,优化各位置区间工艺参数,进而实现管道全位置激光焊接.结果表明,摆动激光束周期性的作用于填充焊丝,产生的反冲压力能够促进熔滴过渡,使得焊丝始终以"液桥"形式向熔池过渡;同时摆动激光增强了熔融金属侧向流趋势,提高熔池界面表面张力,削弱空间多位置下重力对熔池形貌的影响,保证各空间位置熔池均能稳定存在,焊缝成形连续均匀.     

Welding deviation detection method based on weld pool image contour features

Automatic on-line detection of welding deviation based on machine vision is one of the key technologies of arc welding robot tracking welding,in which obtaining high quality weld pool image and accurate welding deviation detection algorithm are two important steps of tracking welding.Through the research and analysis of the weld pool image of gas metal arc welding(GMAW),it was found that the weld pool contains abundant welding information.First,the average gray value of the weld pool image can reflect the interference degree of arc to weld pool image and the heat input of welding process.Secondly,the tip of the weld pool image contour can reflect the center of the groove gap.Finally,the horizontal distance between the center coordinate of the wire contour and the tip coordinate of the weld pool image contour can reflect the welding deviation.On the basis of analyzing the characteristics of weld pool image,this paper proposes a new method of weld seam deviation detection,which includes the collection of weld pool image,image preprocessing,contour extraction and deviation calculation.The results of the tests and analyses showed that the maximum error of the on-line welding deviation obtained was about 2 pixels(0.17 mm) when the welding speed was ≤60 cm/min,and the algorithm was stable enough to meet the requirements of real-time deviation detection for I-groove butt welding.The method can be applied to the on-line automatic welding deviation detection of arc welding robot.     

Weld pool boundary and weld bead shape reconstruction based on passive vision in P-GMAW

A passive visual sensing system is established in this research, and clear weld pool images in pulsed gas metal arc welding （ P-GMA W） can be captured with this system. The three-dimensional weld pool geometry, especially the weld height, is not only a crucial factor in determining workpiece mechanical properties, but also an important parameter for reflecting the penetration. A new three-dimensional （3D） model is established to describe the weld pool geometry in P-GMAW. Then, a series of algorithms are developed to extract the model geometrical parameters from the weld pool images. Furthermore, the method to reconstruct the 3 D shape of weld pool boundary and weld bead from the two-dimensional images is investigated.     

Classification of MAG weld pool image based on moment invariants and fisher

In order to discover characteristics of various kinds of weld pool image and identify a single image,seven image features are extracted to describe the corresponding surface formation quality by the mo...     

Weld pool vision sensing and image processing for GMAW

It is difficult to acquire satisfied weld pool image by CCD sensor during gas metal arc welding(GMAW), for arc disturbs violently, welding current is great and working frequeacy is high. By using CMOS vision sensor to GMA W process, the vivid weld pool image is collected at any time, furthermore, whose gray compression ratio is controllable by sensor hardware circuit developed. Acquired weld pool image is firstly pre-processed by using Wiener filter and Ostu threshold segmentation algorithm. Subsequently separation between weld pool intage and cathode mist region is conducted by means of mathematical morphological algorithm, and the edge of weld pool image is extracted by using Prewitt algorithm.     

基于透红外视觉传感的GMA-AM熔池图像质量评价

针对熔化极气体保护电弧(gas metal arc,GMA)增材制造(additive manufacturing,AM)图像感知,提出一种透红外熔池视觉传感方法. 为客观评价熔池图像质量,综合图像灰度、纹理、形状和频谱等四类特征定义了熔池图像质量评价参数φ. 结果表明,φ值越大,图像质量越好. 透红外熔池图像质量评价参数φ远大于近红外窄带熔池图像,熔池更清晰,对比度更高. 相比800,850和930 nm等透红外滤光片,990 nm透红外滤光片能过滤大部分电弧连续谱和特征谱,获得的熔池信息量最大,对比度更明显,边缘更清晰,细节信息更丰富,熔池图像质量最佳,是最佳取像窗口.     

PCNN based image processing and feature extraction of dual-bypass gas metal arc weld pool

In manual welding process, skilled welders can adjust the welding parameters to ensure the weld quality through their observation of the weld pool surface. In order to acquire useful information of the weld pool for control of the welding process and realizing the automatic welding, the measurement system of DB-GMA W process was established and the weld pool image was obtained by passive vision. Then, three image processing algorithms, Sobel, Canny, and pulse coupled neural network （PCNN） were detailed and applied to extracting the edge of the DB-GMA weld pool. In addition, a scheme was proposed for calculating the length, maximum width and superficial area of the weld pool under different welding conditions. The compared results show that the PCNN algorithm can be used for extracting the edge of the weld pool and the obtained information is more useful and accurate. The calculated results coincide with the actual measurement well, which demonstrates that the proposed algorithm is effective, its imaging processing time is required only 20 ms, which can completely meet the requirement of real-time control.