An extended bi-dimensional empirical wavelet transform based filtering approach for engineering surface separation using high definition metrology

Abstract

Surface topography is an important index to evaluate product quality, and filtering is a key step to realize the extraction of embedded surface components. Based the reported achievements, bi-dimensional empirical wavelet transform (BEWT) outperforms the traditional filtering approaches in nonlinear and non-stationary signal processing. High definition metrology (HDM) is used to measure the engineering surface, which provides a 3D inspection of the entire surface. This paper proposes a novel extended bi-dimensional empirical wavelet transform (EBEWT) based filtering approach for engineering surface separation. The proposed approach consists of HDM data preprocessing, adaptive spectrum segmentation, EBEWT and areal surface evaluation. To verify the performance, the proposed approach is compared with areal Gaussian filter and the original BEWT by numerical simulations and four real-world cases. The results demonstrate that the EBEWT based filtering approach can separate the engineering surface properly, thereby indicating the irregularities of manufacturing process and function behaviors of the part.