一种新型稳健样条轮廓滤波器

Novel Robust Spline Filter for Profile Evaluation

在表面计量学中,通常利用轮廓滤波器提取轮廓中线作为表面评定的基准线。由于传统轮廓滤波器极易受到划痕和深谷等奇异信号影响而产生失真甚至错误的中线,为了增强其稳健性并保证评定精度,国际标准中给出了两种替代滤波器,即稳健型样条滤波器和高斯回归滤波器。然而,这两种滤波器的构造原理和实现算法迥异,其滤波特性差异较大,从而导致对同一表面具有不同的评定结果。为解决这一问题,提出将M估计权函数与样条滤波器相结合构造了一种新型稳健样条滤波器。该滤波器不但具有更强的抗差能力,其计算效率也远高于国际标准推荐的稳健滤波器。更重要的是,大量实验表明,其滤波特性与高斯回归滤波器接近,该滤波器的应用有助于表面评定标准的传递和测量结果比对。

In surface metrology, the profile filters are the most widely used filtering technique to determine the reference line for surface assessment. Because that these filtering mean lines are always distorted by freak characteristics of the scratches and peaks of the profile, the robust spline filter and the Gaussian regression fiher were recommended by ISO standards. However, the different algorithm and the different filtering characteristics lead to different mean lines for the assessment of the same surface. In order to solve the problem, the M - estimation algorithm is introduced into the spline algorithm and a novel robust spline filter is proposed. The new filter can perform a better robust ability than the robust spline filter recommended by ISO standard with lower computation. Furthermore, many mesurement experiments show that the new filter possesses the same filtering characteristics as the Gaussian regression filter, which helps to the international standards transfer and the comparision of the surface evaluation results,