Filtering off noise from a fringe pattern is one of the key tasks in optical interferometry. In this Letter, using some suitable function spaces to model different components of a fringe pattern, we propose a new fringe pattern denoising method based on image decomposition. In our method, a fringe image is divided into three parts: low-frequency fringe, high-frequency fringe, and noise, which are processed in different spaces. An adaptive threshold in wavelet shrinkage involved in this algorithm improves its denoising performance. Simulation and experimental results show that our algorithm obtains smooth and clean fringes with different frequencies while preserving fringe features effectively.
© 2012 Optical Society of America
Instrumentation, Measurement, and Metrology
Original Manuscript: October 6, 2011
Revised Manuscript: December 9, 2011
Manuscript Accepted: December 12, 2011
Published: February 1, 2012
Shujun Fu and Caiming Zhang, "Fringe pattern denoising via image decomposition," Opt. Lett. 37, 422-424 (2012)