齿距视觉测量的齿廓图像边缘失真修正算法  

Algorithm of tooth profile image with pitch measurement for edge distortion correction

作　　者：支珊 赵文珍[1] 赵文辉[1] 段振云[1] 孙禾[2] Zlii Shan;Zhao Wenzhen;Zhao Wenliui;Duan Zhenyun;Sun He(School of Mechanical Engineering,Shenyang University of Technology,Shenyang 110870,China;School of Electrical and Information Engineering,Liaoning institute of Science and Technology,Benxi 117000,China)

机构地区：[1]沈阳工业大学机械工程学院,沈阳110870 [2]辽宁科技学院电气与信息工程学院,本溪117000

出　　处：《电子测量与仪器学报》2019年第4期50-57,共8页Journal of Electronic Measurement and Instrumentation

基　　金：十二五国家科技支撑计划(2014BAF08B01);国家自然科学基金(51275073)资助项目

摘　　要：针对机器视觉测量齿距时因齿廓图像边缘失真导致偶然性测量误差的问题,提出了一种基于统计分离测量误差类型的齿廓图像边缘失真修正算法。首先,对基于Bertrand曲面模型检测齿廓亚像素边缘算法进行改进,获取亚像素齿廓边缘偏距信号;然后,采用阈值法识别包含边缘失真信号的齿廓边缘偏距信号段,利用频数分布统计的方法计算齿廓渐开线的初始相位角;最后,实现齿轮齿距测量。试验结果表明,该算法进行齿廓图像边缘失真修正测得的齿距与经过严格清洗处理后不存在齿廓图像边缘失真时测得的齿距十分接近,单个齿距偏差小于0. 75μm。齿距累计偏差测量结果与M&M3525型齿轮测量中心的测量结果相近,偏差值为2. 60μm。In order to solve the problem of accidental measurement error caused by edge distortion of tooth profile image when measuring tooth pitch by machine vision,an edge distortion correction algorithm for tooth profile image based on statistical separation of measurement error types is proposed. Firstly,improving the algorithm of detecting the sub-pixel edge of tooth profile based on Bertrand surface model to obtain the offset signal of sub-pixel edge. Then,the threshold method is used to identify the offset signal segment of tooth profile edge containing the edge distortion signal,and the frequency distribution statistics method is used to calculate the initial phase angle of the tooth profile involute. Finally,the gear pitch measurement is realized. The experimental results show that the pitch measured by using this algorithm to correct the edge distortion of the tooth profile image is very close to the pitch measured when there is no edge distortion of the tooth profile image after strict cleaning treatment,the deviation of individual tooth pitch is less than 0. 75 μm.The measurement result of the accumulated deviation of tooth pitch is similar to that of the M&M3525 gear measuring center,and the deviation value is 2. 60 μm.

关 键 词：机器视觉 齿距测量 图像边缘失真 Bertrand曲面 误差分离 齿廓定位 

分 类 号：TH161[机械工程—机械制造及自动化] TP391[自动化与计算机技术—计算机应用技术]