摘要
在基于机器视觉的齿轮测量过程中,由于齿轮外轮廓各部分的形状、精度以及测量要求均不相同,需要对齿顶、齿廓和齿根分别进行数据处理,因此齿轮外轮廓分段至关重要。针对机器视觉系统测量视野小,单幅图像只能包含齿轮部分轮齿的问题,提出一种齿轮外轮廓分段方法。通过八邻域边缘追踪法提取齿轮外轮廓边缘,对该边缘进行最小二乘法圆拟合确定半径,并进行直线拟合确定齿轮中心逼近方向,经多次拟合后,逐渐逼近齿轮中心。齿轮中心确定后,实现对齿轮齿顶、齿廓和齿根部分的分段。实验结果表明,该方法可以确定齿轮中心,实现对齿轮外轮廓分段,程序简单,可操作性强,有利于提高系统的测量精度。
In the gear measuring process based on machine vision,the shape,accuracy and measuring requirement of gear outer contour are different,it is necessary to fit edge of tooth top,tooth profile and tooth root separately,so it is important to complete the segmentation of gear outer contour. Aiming at the problem of small field of system measuring vision and capturing picture only contain part of gear tooth,a method of gear outer contour segmentation is proposed,Extracting gear outer contour edge by eight neighborhood edge tracking method,and fitting edge points by least squares circle to determine the radius of the circle,and fitting a straight line to determine the approximate direction of the gear center,after several fittings,gradually approaching the gear center. After determining the gear center,complete the segmentation of gear tooth top,tooth profile and tooth root. The experimental result shows that the method can determine the gear center and comply segmentation of gear outer contour,the program is simple and easy to operate,and it is benefit to improve the measurement accuracy of system.
出处
《组合机床与自动化加工技术》
北大核心
2016年第4期117-120,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
十二五国家科技支撑计划(2014BAF08B01)
关键词
视觉检测
齿轮分段
八邻域追踪
最小二乘法
齿轮中心
vision measurement
gear segmentation
eight neighborhood tracking
least-square fitting
gear center
