摘要
针对滚动轴承滚动体检测的研究较为匮乏且精度较低的问题,文章提出一种将机器视觉应用于轴承滚动体数目和尺寸检测的检测方法。利用小波降噪的方法对相机采集到的图像进行优化,再使用最大类间方差法计算得出适用于降噪后图片的最佳阈值来进行二值化处理,采用形态学运算强化滚动体特征,最后为了消除最小二乘法拟合结果的偏差,利用迭代法改进最小二乘法得出滚动体的数目及准确尺寸。实验结果表明,该检测方法的检测准确性和精度符合工业检测的要求。
Existing studies on the detection of rolling elements for rolling bearings are relatively scarce and have low precision,so this paper presents a method for detecting the number and size of bearing rolling element.Wavelet denoising method is used to optimize the image acquired by the camera.Using the optimal threshold value that calculated by using the most common method of variance to process the binary processing.The characteristics of the rolling body are enhanced by morphological operation.Finally,the number and exact size of the rolling body are calculated by using least squares method which is improved by iterative method.The results of the experiment show that the accuracy and precision of the detection method meet the requirements of industrial inspection.
引文
[1]韩亮.基于机器人视觉的轴承缺珠检测方法[J].机床与液压,2017,45(14):151-153.
[2]杨晓蔚,伍海云.滚动轴承对滚动体的技术要求[J].轴承,2010(8):57-61.
[3]陈琦,阮鸿雁.基于机器视觉的滑动轴承缺陷检测系统设计[J].组合机床与自动化加工技术,2017(5):92-95.
[4]徐平,常英杰,薛凌云.基于边缘像元投影的微小轴承亚像素边缘检测[J].仪器仪表学报,2016,37(10):2300-2306.
[5]卢满怀,范帅,汤绮婷.基于机器视觉的轴承套圈检测系统[J].轴承,2017(5):39-44.
[6]王恒迪,李莎,杨建玺,等.轴承端面缺陷的视觉检测方法[J].轴承,2016(3):60-63.
[7]崔明,顾启民,黄震.基于机器视觉的轴承滚动体缺陷检测算法研究[J].组合机床与自动化加工技术,2015(11):74-78.
[8]耿亮.基于表面特征的钢球分拣技术的研究[D].济南:山东理工大学,2015.
[9]曹风云.基于机器视觉的钢球表面质量分拣系统[D].济南:山东理工大学,2016.
[10]Gao Y,Zhang G Y.High Precision Machine Vision System for Outer Diameter Inspection Bearing[J].Key Engineering Materials,2013,552:313-318.
[11]Liu L J,Wang Y N.Rolling Bearing Outside Diameter Inspection System Based on Machine Vision[J].Journal of System Simulation,2007,19(21):4981-4972.
[12]刘青,张进华,黄军勤.基于机器视觉的钢球表面缺陷检测和分类[J].轴承,2013(10):44-48.