Abstract
We present a modification of the multigrid method for the minimization of quadratic cost functions that result from the formulation of several problems in the digital processing of fringe-pattern images by using the Bayesian estimation-theory framework. With this modification the method can be applied to irregular domains and results in substantial savings in computational cost. We compare it with other state-of-the-art numerical techniques and present examples of its application to image smoothing without edge effects, robust quadrature filtering for the phase demodulation of spatial-carrier fringe patterns, and robust phase unwrapping.
© 1998 Optical Society of America
Full Article | PDF ArticleMore Like This
José L. Marroquin, Mariano Rivera, Salvador Botello, Ramón Rodriguez-Vera, and Manuel Servin
Appl. Opt. 38(5) 788-794 (1999)
Mariano Rivera, Jose L. Marroquin, Manuel Servin, and Ramon Rodriguez-Vera
Appl. Opt. 36(32) 8381-8390 (1997)
Daniel Crespo, Juan Antonio Quiroga, and Jose Antonio Gomez-Pedrero
Appl. Opt. 43(33) 6139-6146 (2004)