Three-dimensional lookup table for more precise SAR scatterers positioning in urban scenarios

Interferometric Synthetic Aperture Radar (InSAR) applications in urban scenarios require higher geometric accuracy than conventional ones. However, precise positioning of radar scatterers within complex urban structures remains a significant challenge. This study introduces a novel three-dimensional lookup table (3D LUT) positioning method for precise pixel positioning in complex urban environments. It addresses the complex geo-coding/radar-coding problems in side-looking SAR imaging by utilizing information provided by the external LiDAR/optical point cloud. We demonstrate that our 3D LUT significantly outperforms the traditional LUT, offering a more detailed and precise transformation between geographic and radar coordinate systems. Our case study shows that 3D LUT provides a high level of positioning accuracy, with nearly zero bias and a standard deviation of less than 0.5 pixel in radar-coding. Conversely, traditional LUTs exhibited substantial offsets and standard deviations in the same case. Meanwhile, the proposed 3D LUT framework paves the way for the precise integration of multi-source geodata and SAR data. It enables multiple applications in SAR imaging and analysis, including precise change detection, enhanced training dataset preparation for SAR deep learning, and improved Multi-temporal InSAR (MT-InSAR) processing. The source code and sample dataset are available at https://github.com/caigenszu/3DLUT.