Improved branch‐cut phase unwrapping strategy based on dynamic adjacent table

Abstract

Phase unwrapping (PU) is a key step of interferometric SAR signal processing. The accuracy of the unwrapping phase directly affects the accuracy of the digital elevation model (DEM). The traditional path-integral methods and global-planning methods cannot take into account both PU accuracy and processing efficiency, and causes PU error diffusion and large-area PU holes. To solve these problems, this study proposed an improved PU strategy based on dynamic adjacent table. For the problem of unwrapping holes that appears in traditional path-integral methods, this study introduces improved local quality maps to guide the path-integral operation, which effectively realises the phase compensation of the PU holes. For the low efficiency of the traditional branch-cut algorithm, this study implements the real-time integration path optimisation planning by further introducing the dynamic adjacent table and greatly reduces the data volume of the adjacent table. Thus, the efficiency of PU was greatly improved. Finally, the proposed method was validated by the airborne dual-antenna interferometric SAR experimental data.