A deterministic descriptive regularization-based method for SAR tomography in urban areas

ABSTRACT

In recent years, Synthetic Aperture Radar (SAR) Tomography (TomoSAR) has ascertained great potential for the three-dimensional (3-D) reconstruction of observed scenes, especially in urban areas. However, the number of proceed snapshots (observations) is usually less than that of slant height samples (unknowns) in TomoSAR inversion processes. This impairs the quality of the reconstructed vertical information. To cope with this issue and improve the reliability of reconstructed vertical information, this paper investigates the possible potential of a deterministic descriptive regularization-based method. Deterministic descriptive regularization is a well-conditioned optimization framework based on the descriptive idea of a regularization solution. This strategy can help to mitigate the effect of the ill-posed problem. Thus, it can assist SAR tomography to deal with the possible impairing issues arising from low numbers and the distribution of baselines. For this purpose, the result of the proposed strategy is compared with the outcomes from the standard TomoSAR techniques, including Beamforming, Capon, and Minimum Norm. The proposed method for reconstruction of the reflectivity function of the observed scene has been performed on a dataset acquired by the Sentinel-1 sensor in 2022 over Tehran City, Iran. The experimental results indicate that the proposed algorithm can estimate building heights with a vertical accuracy of better than 91%. These results demonstrate the great potential of the proposed method for reconstructing the full 3-D images of urban areas.

KEYWORDS:

• synthetic aperture radar tomography (TomoSAR)
• deterministic descriptive regularization-based method
• constrained least square (CLS) and weighted constrained least square (WCLS)
• Tikhonov regularization

Acknowledgements

The authors would like to thank the scientific agency of the United States Geological Survey (USGS) for providing the Sentinel-1 data images. Also, they would like to thank the Remote Sensing Institute of K.N. Toosi University of Technology and the Radar Laboratory of the Faculty of Geodesy and Geomatics Engineering of K.N. Toosi University of Technology.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.