针对飞切法加工的玻璃元件面形的评价, 研究了小波变换在光学表面低频段和中频段误差的分析方法。采用小波分解变换方法将波前信号的多频率成分分离, 选取小波基 bior2.8, 有效地提取出元件的面形信息, 同时将所需的中频段信息与占主成分的低频信息分离, 得到中频段的特征频率为 0.18 mm.1。该方法避免了频域滤波法所导致的幅频特性失真的问题, 且可以更好地分析波前信号的细节和频率特性。

To evaluate the surface of glass elements that achieved by single point diamond turning (SPDT) operation, the wavelet analysis method is employed to analyze surface and surface ripple.In this work, the wavelet analysis method is introduced to decompose and reconstruct the sampled surface profile signals with the wavelet by using bior2.8 software. The surface is extracted effectively and the signals of the middle-frequency is separated from the surface. The characteristic frequency is 0.18 mm.1. The wavelet analysis method avoids the problem of amplitude distortion which exists in the spectral-filtering method. Moreover, the wavelet multiscale analysis method is applied to analyze the details and frequency performance of the wavefront.