基于机器视觉的工件Mark点快速提取方法研究

Research on Rapid Extraction Method of Workpiece Mark Points Based on Machine Vision

在现代化工业生产流程中,工件Mark点类型繁多且提取过程既复杂又低效。为解决这一问题,提出了一种基于机器视觉的工件Mark点快速提取的方法。首先,使用工业相机采集原始图像,然后对图像进行预处理以减少噪声和锯齿的影响。随后,针对目标边缘提取繁琐且不准确的问题,提出一种基于OpenCV的卡尺工具算法,该算法通过在检测边缘周围产生等大等距的测量矩形,利用亚像素图像及改进的Canny边缘检测算子获得测量矩形内的最佳边缘点,再利用基于梯度下降法的最小二乘法来进行直线或圆的拟合,通过该算法可以快速高效地获得工件的直线或圆边缘。最后,通过组合这些直线或圆边缘来获得所需的Mark点。实验证明,该方法可以快速而准确地拟合工件的直线边缘或圆边缘,并最终成功获取到相应的Mark点,定位精度可达到亚像素级,且具有较高的稳定性。

In modern industrial production processes,there are various types of workpiece Mark points,and the extraction process is both complex and inefficient.To solve this problem,a method for fast extraction of workpiece Mark points based on machine vision is proposed.Firstly,industrial cameras are used to capture the original images,and then the images are preprocessed to reduce the influence of noise and aliasing artifacts.Subsequently,addressing the challenges of tedious and inaccurate extraction of target edges,a caliper tool algorithm is introduced based on OpenCV,the algorithm generates equally spaced measurement rectangles around detected edges,utilizes sub-pixel images and an improved Canny edge detection operator are used to obtain optimal edge points within the measurement rectangles.Further,it employs a least squares method based on gradient descent for the fitting of lines or circles.This algorithm allows for the rapid and efficient extraction of straight or circular edges on the workpiece.Finally,the required Mark points are obtained by combining these straight or circular edges.Experimental results demonstrate that this method can rapidly and accurately fit the straight or circular edges of the workpiece and successfully obtain the corresponding Mark points,the positioning accuracy can reach sub-pixel levels,and the method exhibits high stability.