Phase unwrapping: a new max-flow/min-cut based approach

The paper presents a new max-flow/min-cut approach for recovering the absolute phase from modulo-2π phase, the so-called phase unwrapping (PU) problem. The adopted criterion is the minimization of the L^p norm of phase differences, leading to computationally demanding integer optimization problems. The unwrapped phase is computed iteratively through a sequence of binary optimizations, each one mapped onto a max-flow/min-cut problem on a certain graph. Accordingly, we name this new algorithm PUMF (for phase unwrapping max-flow). A set of experimental results illustrates the effectiveness of PUMF approach, namely for 0 < p < 1, where most competitors fail.