A new approach for gaussian filter in surface metrology

Gaussian filter is a standard filter in surface metrology recommended by ISO11562. Its recognized problem is called "end effect". In order to overcome this disadvantage, a new approach based on approximating spline is developed to realize Gaussian profile filter. First, the configuration of approximating spline filter is investigated, and it describes that this filter inherits all the characteristics of ordinary spline filter e.g. no phase distortion and no end effect. Then, the approximating coefficient selection is discussed. This filter can approximate to Gaussian with little 4.36% deviation. Moreover, the approximating deviation can be decreased to less than 1%, when cascaded approximating spline is adopted. It is proved that approximating spline filter not only accords with the transmission characteristics of ISO, but restrains the end distortion on data sequence. So, it is also named spline achievement of Gaussian filter. Finally, the whole computational procedure is illustrated and applied to a workpiece to acquire mean line. The experimental results indicate that the filtering computation of this new algorithm for 11200 data points on general computer only spends 130ms.