研究室のメンバー Fathin Nurzaman（D2） の論文がIEEE Geoscience and Remote Sensing Lettersに採択されました。
https://ieeexplore.ieee.org/document/10685070
Abstract:
This letter proposes a new Atmospheric Phase Screen (APS) compensation method for ground-based Synthetic Aperture Radar Interferometry (InSAR). Conventionally, the APS trend is retrieved employing the assumption that the displacement component is a high spatial frequency component thus making it separable from the APS trend and easy to be identified as outliers during the trend estimation whether using least squares or spectral analysis. However, landslides generally occur over a wide area forming a low spatial frequency component similar to the atmospheric component which could give significant bias during the parameter estimation. This led to the suppression of the observed displacement from its true since it was partly misidentified as the APS. This specific APS misidentification issue is rarely mentioned in previous literature regarding APS compensation. A robust method to address this issue is proposed in this letter, dubbed BeSSaC (Best Sub-Sample Consensus). This method employs a robust estimation technique using the least median absolute residual criteria, it is shown that a wide displacement component can be effectively separated from the APS using this technique.