轴承保持架计算机视觉检测方法研究

The study of bearing cage inspecting adopting computer vision

根据铁路轴承保持架的检测要求,研究了基于图像边缘信息的保持架窗孔梁计算机视觉检测方法。根据检测对象的特点,确定了最佳的保持架边缘检测算法。在算法的具体实现过程中,提出了弦长法确定保持架边缘大端外圆圆心、区域划分检测窗孔梁宽度边缘直线等计算方法,实验表明,将计算机视觉引入轴承保持架生产的在线检测成本低、效率高,检测精度满足要求。

A new computer vision inspecting tec hnology based digital image processing have been re-searched aiming to Railway Bearing C age(RBC)in this paper.The measurement principles of computer vision inspecting system have been analyze d and the system frame has been design ed.The inspecting strategy of RBC windows bars based image edge detection was studied.Aseveral kind of new arithmetic,including chord length to finding center of RBC outline circ le and area mesh detecting edge line o f RBC windows bars,have been studied and calculated with compute r program.The experiment data indic ated that computer vision inspectin g system of RBC is fulfilled the demanding.This method is simple and practi cal.