基于聚类和特征检测的焊缝和钨针定位方法

钢管氩弧焊接过程中需要实时判断钨针针尖与钢管焊缝中心线是否对齐,以控制焊接质量。用肉眼判断误差较大,且效率较低。为此,提出一种基于聚类和特征检测的氩弧焊图像中焊缝中心线坐标和钨针针尖坐标位置计算方法。该方法利用幂次变换的方法增强图像对比度,用直觉模糊C均值聚类方法对图像进行二值化,用Canny算子边缘检测的方法提取图像特征,检测出钢管氩弧焊图像中的熔池、焊缝中心线以及钨针针尖位置。实验结果表明,该方法能够快速准确地定位焊缝中心线和钨针针尖位置。

Location method of weld seam and tungsten needle location based on clustering and feature detection

In the process of argon arc welding of steel pipe,it is necessary to judge whether the tungsten needle tip is aligned with the weld center line of steel pipe in real time,so as to control the welding quality.The error of judgment by naked-eye is large and the efficiency is low.In this paper,a method based on clustering and feature detection for calculation of weld center line coordinates and tungsten needle tip coordinates is presented.In this method,the contrast of the image is enhanced by the method of power transformation,and the image is binarized by intuitionistic fuzzy C-means clustering.The Canny operator edge detection is used for image features.The weld pool,weld center line and tungsten needle tip position in the image of argon arc welding of steel tube are detected.The experimental results show that the proposed method can quickly and accurately locate the center line of weld seam and the position of tungsten needle tip.