基于机器视觉的硅钢板边浪平直度测量方法

针对硅钢生产过程中存在的边浪平直度难以检测问题,提出了一种基于机器视觉的硅钢板边浪平直度测量方法。在辅助光源照射下使用相机采集边浪图像信息,强化边浪的图像明暗特征,提高检测精度的同时实现非接触式测量。同时提出了一种基于轮廓检测的硅钢板边浪平直度算法,首先利用种子点生长法的思想对边浪区域进行轮廓定位;其次对图像进行 3 个尺度的高斯滤波,重建图像,增强轮廓细节;最后基于优化的 Canny 边缘检测方法对边浪实现自适应阈值轮廓提取,计算平直度,从而实现非接触式高精度边浪平直度测量。将平直度测量仪测量结果与该方法结果进行对比,结果表明,所提方法平直度测量误差不超过 0.05% 。

Measurement Method of Edge Wave Straightness of Silicon Steel Plate Based on Machine Vision

Aiming at the problem that the side wave straightness of the side wave is difficult to detect in the production process of silicon steel , a measurement method of side wave straightness of the silicon steel plate based on machine vision is proposed.Under the illumination of the auxiliary light source , the camera is used to collect the image information of the side wave , which strengthens the image light and dark characteristics of the side wave , improves the detection accuracy and realizes non-contact measurement.At the same time , a flatness algorithm of the edge wave of the silicon steel plate based on contour de- tection is proposed , and the contour positioning of the edge wave area is first used by the idea of seed point growth method.Secondly , the image is filtered on three scales , reconstructing the image and enhancing the contour details ; Finally , based on the optimized Canny edge detection method , the adaptive threshold profile extraction is realized for the side wave , and the flatness is calculated , so as to realize the non-contact high-precision side wave flatness measurement.Comparing the measurement results of the flatness measuring instrument with the results of this method , the results show that the measurement error of the flatness of the processed method does not exceed 0.05%.