Generalized Gabor-S transform for fringe pattern analysis
Journal
2016 2nd IEEE International Conference on Computer and Communications, ICCC 2016 - Proceedings
Pages
1912-1917
Date Issued
2017
Author(s)
Tsui, S.-Y.
Abstract
Windowed Fourier transform (WFT) and generalized S-transform (GST) have been widely applied to optical 3D measurement based on fringe pattern analysis. However, both transforms suffer from its own limitations. By combining these two transforms, a new transform called 1D-generalized Gabor-S transform (GGST) is introduced to overcome the limitations. The results show that after adding both a second harmonic term and noise, the minimum averages of absolute phase errors for WFT, GST, and GGST are 0.2571, 1.2578 and 0.1735 respectively. It turns out GGST with ridge-based algorithm has been proved the most robust time frequency algorithm suitable for fringe pattern analysis.
Type
conference paper