基于机器视觉的圆环形零件形位尺寸自动测量  

Measurement of Circular Ring Parts' Form and Position Size Base on Machine Vision

作　　者：刘科文[1] 周平[1] 付斌斌[1] 

机构地区：[1]浙江理工大学信息电子学院,浙江杭州310018

出　　处：《工业控制计算机》2010年第7期1-3,共3页Industrial Control Computer

摘　　要：采用机器视觉技术实现对圆环形零件内外径和同心度的非接触式测量。使用200万像素USB数字摄像机获得零件图像,对图像采用双峰法寻找到合适的阈值将图像二值化,运用轮廓跟踪法找到零件的边缘,对边缘数据采用最小二乘法拟合求出零件的内外径和同心度。对1～2cm外径的零件进行实验,绝对误差小于一个像素,与人工测量值对比,最大误差不超过0.03mm。研究结果表明:在机器视觉测量中,摄像机的像素越大、被测物体尺寸越小,则测量精度越高,所以采用高分辨率的摄像机可以实现对高精度的微小零件测量。In this paper,technology of machine vision is used to measure circular ring parts＇ inside and outside diameter and concentricity.The picture of parts is got by 2 million-pixel USB digital camera.During processing,firstly,using bimodal method to find appropriate threshold to get binary image.Secondly,using contour-tracing method to find edge of the parts;thirdly,us- ing the edge data to get the parts＇ inside and outside diameter and concentricity through least-squares method.Experiment of 1～2cm diameter parts,absolute error is less than one pixel.The largest error is less than 0.03mm compared with the manual measurements.The results show that：In the machine vision measurement,the higher the camera＇s pixel level is and the smaller size of the measured object is,the higher the measurement＇s accuracy.So high level pixel camera can be used to measure high-precision small parts.

关 键 词：零件测量 机器视觉 数字图像处理 最小二乘法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]