Digital signal processing techniques for central fringe identification in white-light interferometry

White-light interferometry (WLI) allows the overcoming of the operating range limitation imposed by conventional systems that use highly coherent light sources. Fibre-optic WLI sensors offer high safety, small weight, and resistance to high temperatures, and are ideal for measurements in hazardous conditions. The noise level is, however, increased as the power-coupling between source and fibre is low and vibrations or spatial mismatches are introduced during scanning. This paper reviews the main techniques for the identification of the central fringe in WLI interference patterns and compares them under a worst-case output signal-to-noise ratio of 20 dB, emphasizing success rate, ease of implementation, computational complexity and memory requirements.