Hough变换曲线拟合的轮毂轴承沟道测量方法

A Measurement Method for Hub Bearing Raceway based on Hough Transform Curve Fitting

针对传统的轴承沟道参数测量方法易受异常点影响的问题,提出了一种利用Hough变换曲线拟合的轴承沟道测量方法。首先利用最小二乘拟合算法快速计算圆弧的参数,然后根据最小二乘算法所得的结果确定Hough变换中参数的遍历范围,以此提高Hough变换的计算效率,同时利用Hough变换结果滤除原始测量数据中的异常点,并将处理后的数据重新用于最小二乘圆弧曲线拟合,以此来消除异常点对圆弧参数精确估计的影响。分别使用仿真和实测实验对算法的有效性进行了验证。实验结果表明,与传统的最小二乘曲线拟合算法相比,具有精度高、鲁棒性好等优点。

We propose a novel measurement method for hub bearing raceway based on the Hough transform curve fitting,which can solve the problem but whose results are easily affected by abnormal points using the traditional measurement methods. First,the arc parameters are calculated using the least squares fitting method,and then,the parameter values are used to determine the scope of search for the Hough transform,thus improving the calculation of the Hough transform. The results obtained with the Hough transform can be used to filter the abnormal points,whose influence on measurement accuracy can thus be eliminated when the least squares fitting method is applied to data processing. Both simulation and experimental results prove the validity of the measurement method. Compared with the traditional least squares fitting methods,our measurement method has the advantages of high precision and good robustness.