Multigrid phase unwrapping for interferometric SAR

Presents a new method for weighted least-squares phase unwrapping. The method is a full multigrid (FMG) technique that applies Gauss-Seidel relaxation schemes to the partial derivatives of the given phase data on smaller, coarser grids and transfers the results to the finer grids. The key idea of the author´s multigrid approach is to maintain the partial derivatives of the phase data in separate arrays and to correct these derivatives at the boundaries of the coarser grids. The resulting multigrid algorithm is up to ten times faster than existing algorithms for weighted least-squares phase unwrapping