The classical Markov Random Field (MRF), used extensively in Image Processing was actually derived from the Ising and Potts models used for describing magnetic phase transitions in materials at the atomic scale. This work upgrades the regularizer from the Potts model to the modern Phase Field Model, which quantitatively describes material structure formation, in order to provide a quantitative physics-based regularizer for solving ill-posed problems in microscopy. Examples of phase field regularized denoising are given.

This presentation is part of Minisymposium “MS11 - Computational Imaging for Micro- and Nano-structures in Materials Science (2 parts)”
organized by: Brendt Wohlberg (Los Alamos National Laboratory) , Jeff Simmons (Air Force Research Laboratory) .

Authors:

Jeff Simmons (Air Force Research Laboratory)

Amirkoshyar Ziabari (Electrical and Computer Engineering; Purdue University)

Jeffrey Rickman (Materials Science and Engineering/Physics; Lehigh University)

Charles Bouman (Purdue University)

Keywords:

bayesian methods, image denoising, image enhancement, image segmentation, inverse problems, statistical inverse estimation methods