Two-Dimensional Minimum Discontinuity Phase Unwrapping Using Pretreatment and Edge Detection

Yuangang Lu,Ting Zhang,Xuping Zhang

DOI: https://doi.org/10.1117/12.807025

2008-01-01

Abstract:Two-dimensional phase unwrapping method is important in a great number of applications. Flynn's minimum discontinuity phase unwrapping algorithm can be used to solve many different kinds of phase unwrapping problems successfully, but its main drawback is that, since it takes the whole wrapped phase image as the scanning areas, it requires a large amount of computations and has low efficiency in searching for the phase's discontinuity within the improving loop search area. To overcome this drawback, a novel minimum discontinuity phase unwrapping algorithm using pretreatment and edge detection is proposed. In the new algorithm, a pretreatment operation is carried out on the wrapped phase image to identify the isolated noisy pixels and exclude them from the search areas. Then, the phase's discontinuity areas within wrapped phase image are detected based on edge detection techniques and taken as the improving loop search areas. By these means, the efficiency of the new minimum discontinuity phase unwrapping algorithm is significantly improved because the required scanning range is cut down to a large extent. This new algorithm is described in detail and tested by using a noisy IFSAR experimental wrapped phase image. Unwrapped results make it clear that the new algorithm works very well and is more effectively than the Flynn's minimum discontinuity algorithm.