摘要
快速地对大批量生产齿轮的误差进行在线测量是智能化生产的本质要求,结合机器视觉技术非接触测量直齿轮的加工误差具有重要的意义和可行性。文中借鉴Zernike矩亚像素边缘检测算法,结合图像直方图的最大类间方差方法对在线拍摄的直齿轮图像进行最佳全局阈值处理,在提高轮齿边缘检测效率的同时降低了误差。针对Zernike矩亚像素边缘像素宽度较粗的缺点,对采集的图像进行形态学滤波等处理细化了边缘,在此基础上设计开发了一系列算法,得到了齿轮的基本参数和齿距偏差,并进行了误差分析。通过计算结果与实测值之间的对比,证明文中提出的采用机器视觉技术测量齿轮的算法精度较高,可以满足实际生产过程直齿轮在线测量的需要。
In the process of intelligent production,errors must be rapidly subject to in-site measurement for the mass production of spur gears. Besides,it’s significant to identify the parameters and errors by means of non-contact measurement in combination with machine vision. In this article,based on the detection algorithm of Zernike moment sub-pixel edge,in terms of image processing,the maximum-between-clusters-variance method is adopted to identify the optimal global threshold,which both improves the efficiency in edge detection and minimizes the errors. Since the Zernike moment sub-pixel edge is coarse,the image is processed by morphological filtering to refine the edge. Then,a series of algorithms are worked out to measure the basic parameters of gears and its pitch deviation,and the analysis is carried out on the errors. The comparison between the calculated results and the measured data shows that the above-mentioned algorithm has such advantages as non-contact,good real-time performance and high accuracy in measurement.
作者
程敏杰
王建文
CHENG Min-jie;WANG Jian-wen(School of Mechanical and Power Engineering,East China University of Science and Technology,Shanghai 200237)
出处
《机械设计》
CSCD
北大核心
2020年第3期19-22,共4页
Journal of Machine Design
基金
国家重点研发计划重点专项资助项目(2016YFF0203000).
作者简介
程敏杰(1995—),女,硕士研究生,研究方向:机器视觉。E-mail:chengminjie723@qq.com;通信作者:王建文(1969—),男,副教授,博士,研究方向:机械工程、现代设计方法。E-mail:wangjianwen@ecust.edu.cn
