Two-dimensional phase unwrapping with a multichannel least-mean-square algorithm

This paper presents a computationally efficient two-dimensional phase-unwrapping method based on a multichannel least-mean-square algorithm. The performance of the proposed method is evaluated by applying phase unwrapping to several simulated very noisy images and to a genuine noisy interferometrical image taken from a five-step phase-shift interferogram obtained from a surface plasmon resonance imaging biosensing experiment. The results confirm that the proposed method is more widely applicable, more computationally efficient, and more robust in the presence of noise than the representative methods presented in this paper. (C) 2004 Optical Society of America.