数据自辅助的PSDS-InSAR迭代三维相位解缠算法

Data Self-Assisted Iterative Three-Dimensional Phase Unwrapping Algorithm for PSDS-InSAR

ES评分 0

DOI 10.14075/j.jgg.2025.10.352
刊名
Journal of Geodesy and Geodynamics
年,卷(期) 2026, 46(8)
作者
作者单位

1 中国科学院精密测量科学与技术创新研究院,武汉,430077;
2 中国科学院大学地球与行星科学学院,北京,100049;
3 中国华水水电开发有限公司,成都,610094

摘要
大相位梯度打破了相位连续性假设,常导致形变量估算结果偏低,已成为制约PSDS-InSAR技术实现精准相位解缠的挑战之一。借助深度学习和外部先验信息辅助的相位解缠算法虽能在一定程度上缓解该问题,但易受模型泛化能力不足和外部信息获取困难等因素限制。针对PSDS-InSAR大梯度相位解缠问题,在逐步三维解缠算法的基础上,提出一种无需外部信息的数据自辅助迭代三维解缠算法。该算法将自身估计的初步形变相位模型作为先验信息,通过迭代估计与分离趋势相位来逐步降低相位梯度,依靠数据集自身信息辅助实现稳健的相位解缠。真实数据测试结果表明,该算法可精准恢复大梯度相位信息,实现高空间覆盖度、高精度与高稳健性的形变监测。
Abstract
The large phase gradient disrupts the assumption of phase continuity, often leading to an underestimation of deformation, which has become one of the key challenges constraining the accurate phase unwrapping of PSDS-InSAR technology. Although phase unwrapping methods assisted by deep learning and external prior information can alleviate this issue to some extent, their performance is often limited by factors such as insufficient model generalization capability and difficulties in obtaining external information. To address the problem of large-gradient phase unwrapping in PSDS-InSAR, we propose a data self-assisted iterative three-dimensional phase unwrapping algorithm that does not require external information based on the stepwise three-dimensional unwrapping algorithm. This algorithm uses the preliminarily estimated deformation phase model as prior information, and iteratively estimates and separates the trend phase to gradually reduce the phase gradient, thereby achieving robust phase unwrapping relying on the inherent information of the dataset. Experimental results using real data demonstrate that the proposed algorithm can accurately recover large-gradient phase information, achieving deformation monitoring with high spatial coverage, high accuracy, and high robustness.
关键词
PSDS-InSAR; 大相位梯度; 逐步三维相位解缠; 迭代三维相位解缠; 数据自辅助
KeyWord
PSDS-InSAR; large phase gradient; stepwise 3D phase unwrapping; iterative 3D phase unwrapping; data self-assisted
基金项目
页码 996-1006
  • 参考文献
  • 相关文献
  • 引用本文

马文静, 郭爱智, 徐昊, 张成枫, 徐紫伊, 王舒杭. 数据自辅助的PSDS-InSAR迭代三维相位解缠算法 [J]. 大地测量与地球动力学. 2026; 46; (8). 996 - 1006.

  • 文献评论

相关学者

相关机构