本发明公开了一种退役轴承端面损伤的视觉检测方法,包括:获取退役轴承,采集退役轴承的端面第一灰度图像;对第一灰度图像进行图像预处理,获取滤波图像;对所述滤波图像进行目标区域获取,获取检测区域以及所述检测区域的目标轮廓;获取所述目标轮廓的圆心坐标以及半径信息,基于所述圆心坐标以及所述半径信息获取所述检测区域的极坐标变换图像;获取所述极坐标变换图像中的缺陷位置;将所述极坐标变换图像进行反极坐标变换,获取所述退役轴承上的所有缺陷。本发明通过极坐标变换,将原本分布于环形区域并非常规意义上水平分布的缺陷投影到直角坐标系中处理,一定程度上减少了对于缺陷阈值分割的难度,能够实现对退役轴承端面缺陷的视觉检测。
The invention discloses a visual detection method for end face damage of retired bearings, including: acquiring retired bearings, collecting the first gray image of the end face of retired bearings; The first gray image is preprocessed to obtain filtered image. To obtain the target region of the filtered image, obtain the detection region and the target contour of the detection region; Obtain the center coordinates and radius information of the target contour, and obtain the polar coordinate transformation image of the detection area based on the center coordinates and the radius information; To obtain the defect position in the polar coordinate transformation image; The polar coordinate transformation image is transformed by reverse polar coordinates to obtain all defects on the retired bearing. The invention, through polar coordinate transformation, projects the defects originally distributed in the ring region but not the conventional horizontal distribution into the rectangular coordinate system for processing, reduces the difficulty of defect threshold segmentation to a certain extent, and can realize the visual detection of the end face defects of retired bearing.