2D discrete high order energy operators for surface profiling using white light interferometry

This paper deals with image processing methods using demodulation techniques in white light scanning interferometery for detecting surface roughness. New methods are applied to scanned x-z images (i.e. the depth and lateral axis). The procedure is based on a local demodulation without the need for knowledge of the precise value of the carrier frequency (i.e. the pulse frequency). 2D higher order operators based on the recent Teager Kaiser non linear function are introduced, and compared for robustness to the presence of noise compared with previous ID demodulation techniques. In order to improve the surface detection, cubic bi-variate splines are also employed.