Cost functions for phase unwrapping with redundant integration of finite differences

Unwrapping is a crucial step when analyzing ground displacements with InSAR techniques. A new formulation of unwrapping as linear integer programming problem was given by Costantini et al. under the name redundant integration of finite differences, which can be thought of as generalizing the Minimum Cost Flow (MCF) approach. An important parameter regarding quality of outcomes for both types of algorithm is the choice of cost function. The results of our tests with simulated and real data for different cost functions and both types of algorithm are presented in this paper. Solvers are based on open source code (LEMON, SYMPHONY).