基于图像处理技术的零件孔位尺寸快速测量方法
2022,30(9):60-66
摘要：为提升现有图像测量技术的工程应用价值，针对车身车间现场薄板类零件孔位测量问题，基于二维图像处理技术提出了一种可直接借助智能手机完成零件孔位大小及间距测量的方法；该方法引入测量基准贴纸实现比例尺功能，并且通过手机镜头去畸变算法提升测量精度；借助Canny边缘检测算子得到待测量孔位外边缘轮廓, 使用形态学算法去除干扰；最后计算孔位外围边缘最小外接矩形长宽及中心连线的距离，得到孔位的几何尺寸；两个实验案例表明提出方法所得孔位大小及间距绝对误差小于0.4 mm，能够作为车间现场测量公差较大孔位工作的一种补充手段。
关键词：零件孔位;图像处理;尺寸测量;最小外接矩形
An Efficient Measuring Method for Hole Size of Parts based on Image Processing Technology
Abstract：In order to improve the engineering application value of machine vision measurement technology and solve the problem of hole position measurement in car body shop, an efficient method based on two-dimensional image processing technology is proposed and used to measure the hole size and hole spacing of parts with the assistance of smart phone. A measurement benchmark sticker is employed to realize the function of engineering scale, and distortion correction algorithm can be used to improve the measurement accuracy. With the assistance of Canny algorithm, the outer edge contour of the hole can be obtained, and the morphological algorithm is used to remove the interference. After that, the minimum circumscribed rectangle of the hole in parts is calculated to obtain the geometric dimension. Experimental results show that the absolute error of hole dimension is less than 0.4 mm, the proposed method can be used as a supplementary mean to measure the hole with large tolerance in workshop.
Key words：hole of parts; image processing; dimension measurement; minimum circumscribed rectangle
收稿日期：2022-03-09
基金项目：
