Adaptive Unscented Kalman Filter Phase Unwrapping Method and Its Application on Gaofen-3 Interferometric SAR Data

Area of Science:

• Geosciences and Remote Sensing
• Signal Processing
• Geodesy

Background:

• Phase unwrapping (PU) is critical for digital elevation model (DEM) generation and surface deformation monitoring using interferometric synthetic aperture radar (InSAR).
• Existing methods face challenges with accuracy and noise, particularly in high-noise and low-coherence regions.

Purpose of the Study:

• To introduce an improved phase unwrapping (PU) method for enhanced accuracy and robustness in InSAR data processing.
• To apply the adaptive unscented Kalman filter (AUKF) for the first time in interferometric image PU.

Main Methods:

• Combines an amended matrix pencil model for phase gradient estimation.
• Utilizes an adaptive unscented Kalman filter (AUKF) with a quality-guided strategy based on heapsort for phase unwrapping.
• Applies a circular median filter for final result refinement.

Main Results:

• The proposed method demonstrates superior accuracy in interferometric phase maps compared to MCF, SNAPHU, RPTPU, and UKFPU.
• The method exhibits enhanced robustness against noise, making it suitable for challenging GF-3 SAR data.
• Validation performed using simulated and experimental GF-3 SAR datasets.

Conclusions:

• The developed PU method significantly improves the accuracy and reliability of InSAR-based DEM reconstruction and deformation monitoring.
• The integration of AUKF and a quality-guided strategy offers a robust solution for processing noisy and low-coherence SAR data.
• This advancement is particularly beneficial for applications utilizing GF-3 SAR imagery.