Advanced processing of optical fringe patterns by automated selective reconstruction and enhanced fast empirical mode decomposition

We present a single-image, automatic, adaptive, local and fast algorithm for low quality fringe pattern processing. Fringe pattern is first decomposed into a set of empirical modes using enhanced fast empirical mode decomposition (EFEMD). Comparing with the fast adaptive bidimensional empirical mode decomposition (FABEMD) method the improvement is made in terms of decomposition time while maintaining the mode extraction quality. Subsequently, noise and bias free interferogram is reconstructed using the automatic selective reconstruction method (ASR). In comparison with recently proposed SR-FABEMD-HS (selective reconstruction using FABEMD method and Hilbert spiral transform) algorithm the ASR-EFEMD is automated, objective and fast - it needs only few seconds for complete fringe pattern processing. The proposed ASR-EFEMD technique has been tested using simulated and experimental data including complex closed fringe patterns with wide spatial frequency range. Its superiority over the SR-FABEMD-HS as well as its robustness to fringe pattern noise, background illumination and modulation defects have been corroborated. Algorithm automation and calculation time reduction combined with efficiency boost are considered as main development novelties and practical features.