Linking persistent scatterers to airborne laser scanning points for identifying real objects reflecting SAR signal

Abstract

The Persistent Scatterers Interferometry (PSI) method enables displacement estimation with millimeter accuracy. However, the uncertain positioning of Point Scatterers (PS) makes it difficult to associate them with real objects in space and hampers the interpretation of the results. This article proposes a methodology to enhance the accuracy of PS positions. The methodology successfully establishes links between PS and real objects by associating them with the most likely candidate points extracted from Airborne Laser Scanning (ALS) point clouds. The selection process for suitable candidates is based on ALS analysis of return number, classification, and geometric features determined by neighborhood analysis. The linking process involves determining global transformation parameters for PSs using the Iterative Closest Point (ICP) algorithm. Then, the nearest neighbor search within the error ellipsoid of the PS positions is performed. Tests conducted demonstrated that this method allows for linking more than 80 % and 65 % of the PS derived from Sentinel-1 and TerraSAR-X mission data, respectively, in both ascending and descending geometries. To validate the obtained results, in addition to the quantitative assessment, a qualitative analysis is performed based on a developed 3D visualization module showing all stages of the proposed methodology.