Adaptive filtering of white-light interferometry fringe patterns

Tests are carried out of an on-line adaptive filtering scheme based on the least-mean-square (LMS) algorithm and are specifically aimed at enhancing t he performance of fringe order identification in white-light interferometry (WLI) over a wide range of signal-to-noise ratios (SNR's). Additional capa bility is also provided in accurately pinpointing the central fringe with a minimum resolution of 1/13 of a fringe width anywhere over the middle 1000 pixels of a 1024-pixel CCD array. Computer simulations and subsequent meas urements have demonstrated the superiority of this scheme over the less eff icient centroid-based method previously reported.