Variational image decomposition for automatic background and noise removal of fringe patterns.

In optical metrology, state of the art algorithms for background and noise removal of fringe patterns are based on space-frequency analysis. In this Letter, an approach based on variational image decomposition is proposed to remove background and noise from a fringe pattern simultaneously. In the proposed method, a fringe image is directly decomposed into three components: a first one containing background, a second one fringes, and a third one noise, which are described in different function spaces and are solved by minimization of the functional. A simple technical process involved in the minimization algorithm improves the convergence performance. The proposed approach is verified with the simulated and experimental fringe patterns.