PHASE-SHIFTING INTERFEROMETRY AND MAXIMUM-LIKELIHOOD-ESTIMATION THEORY - II - A GENERALIZED SOLUTION

A novel means of quantitatively assessing the performance of a phase-s hifting interferometer is further investigated. We show how maximum-li kelihood estimation theory can be used to estimate the surface profile from the general case of M noisy, phase-shifted measurements. Monte C arlo experiments show that the maximum-likelihood estimator is unbiase d and efficient, achieving the theoretical Cramer-Rao lower bound on t he variance of the error. We then use Monte Carlo experiments to compa re the performance of the maximum-likelihood estimator with that of tw o conventional algorithms. (C) 1998 Optical Society of America OCIS co des: 120.0120, 120.3180, 100.2650, 000.5490.