A Factorized Variational Technique for Phase Unwrapping in Markov Random Fields

TL;DR

This paper introduces a novel factorized variational approach for phase unwrapping in Markov Random Fields, effectively modeling the zero curl constraint and improving inference in complex imaging scenarios.

Contribution

It proposes a new mean field inference method for phase unwrapping that incorporates the zero curl constraint within a Markov network framework.

Findings

Outperforms least squares method on synthetic data

Successfully applied to large synthetic aperture radar images

Demonstrates improved accuracy in phase unwrapping tasks

Abstract

Some types of medical and topographic imaging device produce images in which the pixel values are "phase-wrapped", i.e. measured modulus a known scalar. Phase unwrapping can be viewed as the problem of inferring the number of shifts between each and every pair of neighboring pixels, subject to an a priori preference for smooth surfaces, and subject to a zero curl constraint, which requires that the shifts must sum to 0 around every loop. We formulate phase unwrapping as a mean field inference problem in a Markov network, where the prior favors the zero curl constraint. We compare our mean field technique with the least squares method on a synthetic 100x100 image, and give results on a 512x512 synthetic aperture radar image from Sandia National Laboratories.<Long Text>