中文摘要：

S变换可以看作是介于小波变换与窗口傅里叶变换之间的变换, 具有很强的时频分析能力, 它将一维信号变换为时间（空间）和频率的函数, 称为瞬时S变换谱, 在沿窗口移动方向上, S变换谱的叠加可得到全局信号的傅里叶频谱。在S变换用于条纹解调时, 局部基频的正确提取是确保获得全局信号基频分量的关键。为此研究了不同的滤波过程对S变换解调条纹相位的影响, 利用不同的滤波器, 在对局部S频谱进行加权滤波后, 叠加局部基频, 得到全局基频分布, 然后再利用逆傅里叶变换获得条纹的相位分布, 从而重建被测物体的面形。讨论了阈值滤波、平顶高斯和平顶汉宁滤波、“脊”线拟合后的平顶高斯和平顶汉宁滤波在S变换轮廓术中的应用, 通过计算机模拟和实验, 初步对比了滤波效果。

英文摘要：

The S-transform, an intermediate step between the wavelet transform and the windowed Fourier transform, is a strong tool to analyze the non-stationary signals, which can be used to transform the local component of a one-dimensional signal into a function with time and frequency parameters, and furthermore, the whole Fourier spectra of the signal can be gotten by the average operation over the time axis. Designing suitable filters to extract the local fundamental component of the S transform spectra are a key part in fringe analysis based on S transform. Several filtering procedures are designed to obtain the useful local fundamental spectrum from S transform spectra, including the threshold filtering, the flat-top Hanning filtering and Gaussian filtering before and after ridge fitting processing, respectively. Averaging the filtered fundamental S transform spectra, the whole Fourier transform fundamental spectra of the tested signal can be obtained, which can be used to reconstruct the surface distribution of the tested object. Computer simulations and experiments are employed to compare the results of these filtering processing.