A robust white-light interference signal leakage sampling correction method based on wavelet transform.

A novel white-light interference signal leakage sampling points (LSPs) correction method based on Morlet wavelet transform and phase seven-step phase shift algorithm (PSA) is presented in this paper. During the correction process, the method can eliminate the LSPs error accurately without changing traditional optical system. Meanwhile, it can effectively de-noise the interference signals and calculate a relatively smooth recovery shape. • The main advantage of the LSPs correction method is that we use wavelet transform to find LSPs in white-light interference signals innovatively for its characteristic of the wavelet coefficient at the distortion points position are different from the general positions, which is different from the traditional method of using wavelet transform for noise reduction and ZOPD extraction. • Furthermore, the algorithm can be applied for any traditional optical system of white-light interferometry. It is simple, accurate and easy to optimize other algorithms. White-light scanning interferometry has become a well-established tool for measuring the microscopic three-dimensional morphology of objects. However, it is sensitive to external disturbances and suffers from leakage sampling, which compromises the accuracy of the measurements. To avoid the error, a robust correction method is proposed, based on wavelet transform, to find leakage sampling point (LSP), and the cubic spline interpolation method is used to complete the interference signal to achieve the signal correction. Further, phase evaluation at each pixel provides a high vertical resolution to characterize the microstructure. In addition, the proposed method exhibits a good denoising performance. Simulations and comparison experiments demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]