Curve-welding-seam three-dimensional reconstruction method based on line structure light vision detection

Abstract

The invention provides a curve-welding-seam three-dimensional reconstruction method based on line structure light vision detection. By using the method, various kinds of welding seams can be effectively detected, reliability is high, the method is simple and detection efficiency is high. A mechanical arm and a line structure light vision system are included. A controller controls and is connected to the mechanical arm. The line structure light vision system comprises a camera and a line structure light projector. The camera is fixed to a tail end of the mechanical arm. The line structure light projector is fixed to one side of the camera and is projected to a surface of a workpiece to be welded. The curve-welding-seam three-dimensional reconstruction method comprises the steps of system calibration, welding seam detection and welding seam reconstruction so that a complete three-dimensional curve welding seam is reconstructed.

Description

technical field [0001] The invention relates to the technical field of computer vision measurement, in particular, it is used to realize weld detection and three-dimensional reconstruction under the guidance of vision, and provides a basis for automatic welding of robots, specifically a three-dimensional reconstruction method of curved welds based on line structured light visual detection. Background technique [0002] In the field of intelligent manufacturing, it is often necessary to obtain three-dimensional information of industrial manufacturing objects. The existing three-dimensional detection devices are divided into two types: contact type and non-contact type. The contact type detection device is not efficient and is prone to scratches on the surface of the object. Therefore, in recent years, non-contact three-dimensional detection devices represented by machine vision have been researched and applied more, and through research, welding robots can be equipped with mac...

AI Technical Summary

Problems solved by technology

[0002] In the field of intelligent manufacturing, it is often necessary to obtain three-dimensional information of industrial manufacturing objects. The existing three-dimensional detection devices are divided into two types: contact type and non-contact type. The contact type detection device is not efficient and is prone to scratches on the surface of the object. Therefore, in recent years, non-contact three-dimensional detection devices represented by machine vision have been researched and applied more, and through research, welding robots can be equipped with machine vision to automatically detect welds, and in visual calibration Based on the three-dimensional detection of welds, it can provide the basis for automatic welding of robots

[0003] With the increasing application of welding robots, it is of great significance to study the visual detection and three-dimensional reconstruction of curved welds to realize automatic welding of robots. To obtain the three-dimensional information of the weld, stereo vision is required. Commonly used stereo vision systems include binocular vision and structured light vision. The detection methods for welds are mainly: (1) Using monocular and structured light vision to detect welds , which projects structured light onto the weld seam, and uses a monocular camera to capture images of the weld seam to realize vision-guided welding; (2) uses binocular vision to detect weld seam information under natural light conditions, and is aimed at welding with small curvature The curve is divided into several sections, and each section is approximated by a straight line to obtain the trajectory of the weld; (3), using line structured light and two cameras to form a stereoscopic vision, and using a 3D reconstruction algorithm to calculate the 3D coordinates of the weld feature points; but Most of the above methods are aimed at straight line welds or small curvature welds, but for complex welds with large curvature, they have not been fully studied. If binocular vision is used, it is necessary to match the two images, which increases the weld detection. The complexity of the algorithm; or real-time welding guided by vision, but due to the existence of arc light, spatter, electromagnetic interference, etc., it affects the reliable detection of the weld, so the weld efficiency is not high

Images