摘要
针对钢领内圈圆度人工接触式检测精度不稳定、效率低、主观误差较大等问题,提出一种基于机器视觉的钢领内圈圆度检测方法。通过建立钢领内圈圆度的机器视觉检测平台,获取背光成像下的钢领图像。利用Sobel算子对钢领内圈边缘进行像素级粗定位,在此基础上通过最大类间方差法确定钢领图像前景与背景的最佳阶跃阈值,将其作为Zernike矩边缘模型的判定条件,提取钢领内圈的亚像素边缘点。采用最小二乘法得到钢领内圈亚像素边缘的理想圆圆心,基于钢领图像信息建立钢领内圈圆度的数学模型,实现钢领内圈圆度的非接触式测量。实验结果表明:对型号PG1-4554的钢领内圈圆度进行检测,得到的最大偏差为0.004 mm,传统方法所得最大偏差为0.007 mm,准确度提升了42%。
To solve the problems of unstable accuracy,low efficiency and large subjective error of manual contact detection for the roundness of steel collar inner ring,a detection method based on machine vision was proposed.By establishing a machine vision detection platform for the roundness of steel collar inner ring,the image of steel collar under backlight imaging was obtained.The edge of the inner ring of the steel collar was roughly positioned at the pixel level by Sobel operator.On this basis,the optimum step threshold of the foreground and background of the image of the steel collar was determined by the maximum interclass variance method,which was used as the judgment condition of Zernike moment edge model to extract the sub-pixel edge points of the inner ring of the steel collar.The ideal center of the sub-pixel edge of the collar inner ring was obtained by the least square method,and the mathematical model of the roundness of the collar inner ring was established based on the image information of the collar,and the non-contact measurement of the roundness of the collar inner ring was realized.The experimental results show that the maximum deviation of the roundness of the inner ring of the steel collar PG1-4554 is 0.004 mm,and the maximum deviation of the traditional method is 0.007 mm,and the accuracy is improved by 42%.
作者
金守峰
焦航
JIN Shoufeng;JIAO Hang(College of Mechanical and Electrical Engineering,Xi′an Polytechnic University,Xi′an,Shaanxi 710600,China;Xi′an Key Laboratory of Modern Intelligent Textile Equipment,Xi′an Polytechnic University,Xi′an,Shaanxi 710600,China)
出处
《毛纺科技》
CAS
北大核心
2022年第4期83-88,共6页
Wool Textile Journal
基金
陕西省重点计划研究项目(2020GY-172)
