汽车第3代轮毂轴承滚动体数目缺失检测

Research on the Missing Detection of the Third Generation Hub Bearing Rolling Element of Automobile

针对第3代轮毂轴承单元在芯轴装完滚子工艺中出现的滚子数目缺失的问题,采用机器视觉方法,研究滚动体缺失在线自动检测。首先,用区域生长法快速获得优质二值图,并用圆形Hough变换对二值图进行圆检测,确定圆心后提取特征圆环;然后,用大津法进行二值化处理,得到滚子的特征图;最后,用上位机封装的计数程序进行训练检测滚动体斑点数,确定滚子缺失的数目。对上述检测方法验证其快速性和稳定性,结果表明,完成一次检测的时间集中在12 s左右,符合生产节拍的要求;同时抽取了80个完好轴承和50个缺失轴承进行检测,识别率分别为92.5%和92%,即该机器视觉系统能够稳定进行工作。该方法能够满足第3代轮毂轴承的生产要求。

In view of the lack of the number of rollers in the 3 rd generation wheel hub bearing unit in the mandrel loading roller process, the machine vision method was adopted to study the on-line automatic detection of missing rolling elements.Firstly, the regional growth method is used to rapidly obtain high quality binary maps, circular detection is performed using a circular Hough transform, the circle center is determined and the feature circle is extracted, then, the Otsuka method is used for binarization to obtain the characteristics of the roller. Finally, using the upper computer package counting program for training to detect the number of rolling body spots, determine the number of roller missing. The above test method was used to verify its rapidity and stability. The results showed that the time to complete one test was concentrated in about 12 s, which was in accordance with the requirements of the production cycle. At the same time, 80 intact bearings and 50 missing bearings were selected for detection, and the recognition rates were respectively 92.5% and 92%, that is, the machine vision system can work stably. This method can meet the production requirements of the third generation of hub bearings.