Precision surface metrology using a rapid optimization method in diffraction phase microscopy

For nanoscale measurement of the surface profile of a test sample, optical interferometry has emerged as a prominent approach in precision metrology. The fundamental requirement in this approach is reliable measurement of phase from the recorded interferogram. However, phase retrieval is constrained by several challenges such as noise, nonuniform amplitude fluctuations, and high computational cost. In this paper, we present a graphics processing unit-assisted second-order optimization technique for demodulating interferograms corrupted by severe noise and uneven fringe amplitude variations. The effectiveness of the proposed method for phase estimation is demonstrated using rigorous numerical experiments that involve varying noise levels and different fringe amplitude fluctuations. Our results show that the proposed method offers significant advantages, including high computational gain and robust phase estimation accuracy, in processing interferograms corrupted with severe noise. Additionally, the practical utility of the method is showcased through its application in measuring surface profiles using experimental interferograms captured via diffraction phase microscopy.