基于机器视觉的齿轮参数测量系统  

Gear Paramete Measurement System Based on Machine Vision

作　　者：陈朝伟 范芳蕾[1] 黄伟莉[1] Chen Zhaowei;Fan Fanglei;Huang Weili(College of Mechanical and Electronic,Donghua Polytechnic University,Nanchang 330013,China)

机构地区：[1]东华理工大学机械与电子工程学院,南昌330013

出　　处：《机电工程技术》2023年第5期113-116,共4页Mechanical & Electrical Engineering Technology

摘　　要：针对传统的机械接触式齿轮测量仪器操作复杂、易产生主观误差等问题,提出一种基于机器视觉齿轮参数非接触式测量系统。首先利用图像灰度变换、直方图均衡化方法对图像进行预处理,提出小波变换模极大值法计算图像的矢量模值和相角,寻找梯度方向极大值点,选取自适应阈值方法提取目标图像边缘。对齿轮的边缘轮廓形状特征,采用最小二乘法拟合原理,计算齿轮的齿根圆直径、齿顶圆直径、齿数和模数等参数。以两种参数的齿轮零件测量为例进行实验分析,结果表明,测量值满足测量精度要求,证明了所提测量系统的可行性。Aiming at the problems of traditional mechanical contact gear measuring instruments,such as complicated operation and easy to produce subjective errors,a non-contact gear parameter measuring research system based on machine vision was proposed.Firstly,the image was preprocessed by means of gray scale transformation and histogram equalization.The wavelet transform modulus maximum method was proposed to calculate the vector modulus and phase angle of the image,find the maximum point in gradient direction,and select the adaptive threshold method to extract the edge of the target image.According to the shape characteristics of the edge profile of the gear,the least square fitting principle was adopted to calculate the parameters such as the diameter of the tooth root circle,the diameter of the tooth top circle,the number of teeth and the modulus of the gear.Taking the measurement of gear parts with two parameters as an example,the experimental analysis was carried out.The measurement results show that the experimental measurement values meet the measurement accuracy requirements,which indicates the feasibility of this measurement method.

关 键 词：机械视觉 非接触式测量 齿轮参数测量 小波变换模极大值 最小二乘法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]