激光焊接焊缝质量视觉检测系统

Visual Inspection System for Laser Welding Seam Quality

研究了LabVIEW在结构光视觉焊缝质量检测方面的应用,提出了一种快速高效的、用于自动化焊接系统图像处理的算法,并开发了激光焊缝质量监控系统。首先使用背景差分算法清除非均匀光照造成的影响,利用类间方差法求解最佳阈值,最大限度地保留了焊缝结构光条纹区域的有效信息;接着设计了自适应形态学收缩算法,使焊缝条纹区域在每列上收缩至一个像素点,从而得到一个没有分叉的骨架;最后利用骨架信息,结合数据拟合和特征参数提取,实现了焊缝的三维重建。试验结果证明,视觉检测系统以及焊缝特征的测量结果令人满意。

The application of LabVIEW in visual quality inspection for structured light welding seam is researched, and the high speed and high efficient image processing algorithm used for automatic welding system is proposed, as well as the laser welding seam quality monitoring system is developed. Firstly, by using background differential algorithm, the influence of the imbalanced illumination is eliminated, and optimal threshold is solved with the class variance method, the maximum effective information for the structured light stripe area of welding seam is retained. Then, the adaptive moiphology contraction algorithm is designed to make the structured light stripe area shrink to a pixel point in each column; thus the skeleton without furcation is obtained. Finally, by adopting the information of skeleton and combining with data fitting and characteristic parameter extraction, the 3D reconstruction of the seam is implemented. The test results verify that the vision detection system and the measurement of the seam features are satisfactory.