An improved minimum discontinuity phase unwrapping algorithm with preprocessing

Phase unwrapping is a very important process in the digital elevation model(DEM) rebuilding of the imaging area from its interferometric phase data. This paper presents a fast phase unwrapping algorithm based on minimum discontinuity optimization, which greatly enhances the efficiency compared with the original algorithm proposed by Flynn. To accelerate the optimization process, a preprocessing process is used to get an initial wrap count by the least-squares method. For that the unweighted least-squares problem can be solved by FFT-based method, the initial wrapped count can be obtained quickly. Then the middle unwrapped result can be obtained by the initial wrap count, and the circle canceling method is used to remove the remaining discontinuities in the middle unwrapped result. Tests with simulated and real data verify the accuracy and efficiency of the proposed algorithm.