科研之友
微信
新浪微博
Facebook
分享链接
摘要
The atmospheric effect is one of the most important limiting factors to Interferometric Synthetic Aperture Radar (InSAR) measurements. It can seriously degrade the quality of InSAR measurements or even render the technology unusable. It is of great importance to develop methods to model and mitigate this effect. A novel method for modeling and estimating the atmospheric phase in SAR interferograms is proposed. The method starts with the spatial and temporal analysis of the characteristics of the atmospheric effects in InSAR. Based on this, it constructs relevant models to describe the stratification and turbulent mixing of atmospheric effects, respectively. Then, it develops the method of robust estimation to determine the model parameters of the stratified atmospheric components, and the method of Matern variogram model-based Kriging interpolation to estimate the parameters of turbulent atmospheric components. Finally, it applies the developed method to estimate and correct the atmospheric effects in InSAR measurements. The proposed method is validated with one ASAR pair over the Yima area in Henan Province. The results show that after the atmospheric correction, the root mean square error of the differences between the InSAR-reconstructed and the reference DEM reduce from 19. 5 m to 5. 3 m, representing an improvement by 72%. In addition, after the correction, the sign of the line-of-sigh (LOS) range change in a mining area varies from positive to negative, indicating that subsidence rather than uplift happening in this area. The corrected interferogram much better reveals the development of the "subsidence bowl" in the mining area. This paper developed a novel method for modeling and mitigating atmospheric effects in InSAR. The method fully exploits the spatial characteristics of atmospheric effects in InSAR, and considers the stratified and the turbulent mixing atmospheric effects as well. Besides, it adopts the Matern model rather than the traditional variogram model in the turbulent mixing modeling. So it works very well. Future work will focus on validating the method in different study areas.
- 出版日期2015-7
- 单位中南大学
