一种线激光V形焊缝中心点识别的改进算法

An improve method of extracting center point of V shape weld based on line-laser

在机器人自动化焊接过程中,准确识别、提取出焊缝位置,对于焊接的精度与质量有着很大的影响。基于线激光的V形焊缝轮廓识别算法,在进行阈值分割时,有时由于线激光与V形焊缝接触时会产生激光的变形,导致局部光强下降,部分轮廓会消失。提出一种改进算法,首先通过双边滤波方法,达到图像平滑的目的;然后通过Canny边缘检测算法,准确地提取出线激光在焊缝上变形的区域;通过膨胀算法处理后,再去除焊缝轮廓留下的干扰直线;再通过直线细化算法,识别出与线激光相交的焊缝边缘轮廓点与中心点。最后识别出焊缝中心点位置。

In the automatic welding process,the accurate identification and extraction of the welding line has a great influence on the precision and quality of welding process. For the recognition algorithm based on line laser,the deformation of laser will lead to a decline in light intensity due to the line contact with the weld in the threshold algorithm. In this paper,an improved method to extract the center point of V shape weld based on line-laser is presented. Firstly,the Bilateral Filter method was used to smooth the image,which helps to keep the contour of laser line while removing the noise. Then the deformation zone of laser on the V shape weld was extracted by Canny Edge Detection algorithm. The Eroding algorithm was used to change the Structure Light on the weld to a continuous solid closed poly line. The interference line left by the contour of the weld was removed. By the line thinning algorithm,the intersection of laser with the edge contour and center line of the weld was detected. Finally,the position of the center point of the weld was identified by the Line Fitting method.