A combined phase unwrapping algorithm for InSAR interferogram in shared memory environment

Abstract

Phase unwrapping is a key step during the signal processing of interferometric synthetic aperture radar (InSAR) data. The precision and efficiency are two key problems of phase unwrapping. The path-following phase unwrapping algorithm usually has high computational efficiency, but low quality areas are prone to have accumulated errors. The global optimization phase unwrapping algorithm can effectively suppress the error propagation by introducing weights, but has low efficiency. This paper presents a combined phase unwrapping algorithm based on quality guided phase unwrapping and minimum discontinuity phase unwrapping algorithm, which improves the precision of the unwrapped result and greatly enhances the efficiency of unwrapping. First of all, according to phase quality map the wrapped phase image is divided into high and low quality areas, and the quantized quality-guided phase unwrapping algorithm is used to obtain the initial unwrapped results. Then low quality areas are grouped to balance the task distribution of low quality area's optimization. Finally, in shared memory parallel computation environment, grouped low quality areas are optimized by the minimum discontinuity phase unwrapping algorithm simultaneously using OpenMP parallel programming model. The tests performed on real InSAR interferogram show that the proposed method is effective.

Under-Sampled Phase Retrieval of Single Interference Fringe Based on Hilbert Transform

“…In the practical application, the interference pattern generated by InSAR technology is similar to the situation [28]- [32] in this paper. Because of the sudden occurrence of clutter, surface subsidence and debris flow in the actual terrain, the under-sampled problems of local interference fringes often appear.…”