基于机器视觉的微间隙焊缝识别方法

A Recognition Method of Micro-Gap Weld Seam Based on Machine Vision

影响焊缝跟踪精密性的关键因素是对焊缝位置的准确识别。利用磁光成像技术采集微间隙(0~0.1 mm)焊缝图像,对焊缝图像的特征进行分析,获得包含焊缝位置信息的感兴趣区域(ROI),针对感兴趣区域进行预处理和边缘检测,能提高图像处理的效率,又根据焊缝磁光边缘方向不一、细节丰富的特点引入多方向上的结构元素对形态学边缘检测算子进行优化,并与梯度边缘检测算子进行比较。结果表明,该方法能高效、准确地检测出焊缝中心的位置,为后续的焊缝跟踪提供有用信息。

The key factor that affects the precision of seam tracking is the accurate identification of weld position. Using the magneto-optical imaging technology to acquire the image of micro-gap weld which is less than 0.1mm, and analyzing the characteristics of the weld seam image, and then obtaining the region of interest （ROI） that containing location information of the weld seam. Image preprocessing and edge detection are conducted in the region of interest, which can improve the efficiency of image processing. And according to the edge characteristics of multidirectional and full of details in the magneto optic image, bringing the structure elements in different directions to optimize the morphological edge detection operator, and compared with the gradient edge detection operators. The results show that this method can effectively and accurately detect the position of the weld center, and provide useful information for the seam tracking.