Adaptive analysis of optical fringe patterns using ensemble empirical mode decomposition algorithm.

An approach based on a novel technique, called ensemble empirical mode decomposition, is proposed to adaptively reduce noise and remove background intensity from a two-dimensional fringe pattern. It can solve the mode-mixing problem of the original empirical mode decomposition caused by the existence of intermittent noise in fringe signals. Then a strategy is developed to automatically identify and group the resulting intrinsic mode functions for the purpose of eliminating noise and background of the fringe pattern. This approach is applied to process the simulated and practical fringe patterns, compared with Fourier transform and wavelet methods.