摘要

Synthetic Aperture Radar Tomography (Tomo-SAR) is an emerging experimental %26quot;coherent data combination%26quot; mode allowing unprecedented full 3-D imaging of complex urban and infrastructure scenarios with layover (%26quot;garbled%26quot;) scatterers, exploiting multibaseline interferometric SAR data stacks. Various approaches have been proposed to improve Fourier-based Tomo-SAR elevation beamforming which is affected by unsatisfactory height sidelobe behaviour and resolution, due to the typical low number of baselines with irregular distribution. Among these approaches, height superresolution multilook beamforming techniques proved to posses interesting capabilities, at the cost of operation with reduced horizontal resolution. In this work, a recently proposed knowledge-based baseline interpolation and the Capon and MUSIC superresolution methods are integrated in to a new Tomo-SAR processor able to offer at a low computational burden height superresolution and sidelobe cleaning with single-look data, allowing full resolution operation, as important in urban and other man-made areas. Results are reported with real ERS data.

  • 出版日期2013-4