Research And Development Of Weld Tracking System Based On Laser Vision

With the introduction of intelligent manufacturing at the national level,automation and intelligence will be the future development direction of industry.Welding is a very important processing means in industrial production,and the realization of welding automation is the inevitable trend of the development of welding technology.Image processing algorithm is applied to the detection field by machine vision technology;the detection precision is greatly improved.The combination of machine vision technology and welding technology,welding seam can be detected in real time,the working efficiency and welding quality can be improved,the damage caused by arc,smoke and dust in the welding process can be avoided,the labor intensity of welding workers can be reduced.In this paper,aiming at the shortcomings of low real-time,low applicability and low welding precision of automatic welding system,a seam tracking system based on laser vision is designed.Laser stripe is formed by transmitting line laser to the surface of the welded workpiece.The weld image is collected by camera;the weld information in the laser stripe is detected by the image processing algorithm,the feature point of the weld is identified and the welding torch is the correct welding position at all times is ensured.The hardware composition of the laser vision system is designed,and the corresponding model is selected according to the actual demand.The visual system is calibrated to get the transformation relationship between the world coordinate system,the camera coordinate system,the image coordinate system and the imaging plane coordinate system,the internal and external parameters of the camera are obtained,and the distortion of the camera is eliminated so as to improve the quality of the weld image.The hand eye calibration is carried out,and the transformation matrix of the camera coordinate system and the terminal coordinate system of the robot is obtained.The Gaussian filter and histogram equalization is used to deal with the weld image,and the noise information in the image is reduced and the information of the laser stripe is enhanced.The local adaptive threshold segmentation is used to segment the image so as to obtain a binary image,and the information of laser stripe and the background are separated.The Central Line of laser striation is extracted with the improved upper and lower average method and Hough transform.Based on the least square method,the accurate center line equation is obtained and the simultaneous equation is solved to solve the intersection point.According to the working flow of the seam tracking system,the control system is compiled with MFC,Open CV and Halcon,and the human-machine interface of the system is friendly.Through experimental verification,the detection error of the welding seam tracking system designed in this paper is within 1.2mm and takes 200 ms.