基于机器视觉的螺纹在线检测方法  

On-line Thread Detection Method Based on Machine Vision

作　　者：牛森涛 赵凤霞[1] 姜菲菲 张琳娜[1] NIU Sentao;ZHAO Fengxia;JIANG Feifei;ZHANG Linna(School of Mechanical and Power Engineering,Zhengzhou University,Zhengzhou 450001,China)

机构地区：[1]郑州大学机械与动力工程学院,郑州450001

出　　处：《机械设计与研究》2021年第6期170-173,183,共5页Machine Design And Research

基　　金：国家重点研发计划项目(2017YFF0206501-01)。

摘　　要：针对汽车制动管螺纹联接零件在线检测需求,提出了一种基于机器视觉的螺纹参数在线检测方法。基于图像矩、仿射变换、双线性插值法对螺纹图像进行倾斜校正,利用中值滤波抑制螺纹边缘区域颗粒化现象。在Canny算法提取的像素边缘基础上,用二次曲线拟合法得到了螺纹图像亚像素边缘,用三次B样条曲线对亚像素边缘进行拟合,根据拟合曲线的曲率检测螺纹角点,最后基于最小二乘法计算出螺纹各尺寸参数。实验结果表明,本文所提出的螺纹检测方法具有效率高、精度高等特点,可以满足在线检测需求。Aiming at the on-line inspection requirements of the threaded connection parts of automobile brake pipes,an on-line inspection method of thread parameters based on machine vision is proposed.Based on image moments,affine transformation,and bilinear intepolation,the thread image is skew-corrected,and the median filter is used to suppress the graining of the thread edge area.Based on pixel edges extracted by the Cany algorithm,the sub-pixel edges of the thread image are obtained by the quadratic curve fiting method,the cubie B-spline curve is used to.fit the sub-pixel edge,and the thread corners are detected according to the curvature of the ftted curve.Finally,dimensions of the thread are determined based on the least squares method.The experimental results show that the thread detection method proposed in this paper has high efficiency and high precision,which can meet the needs of online detection.

关 键 词：螺纹参数 机器视觉 图像矩 亚像素边缘定位 角点检测 

分 类 号：TH166[机械工程—机械制造及自动化] TB921[一般工业技术—计量学]