Improved resolution image focusing by multi-pass ENVISAT ASAR data

Gilda Schirinzi(1) , Gianfranco Fornaro(2) , Vito Pascazio(3) , and Francesco Serafino(2)

(1) Universita' di Cassino, via Di Biasio 43, 03043 Cassino (FR), Italy
(2) CNR-IREA, via Diocleziano 328, 80100 Napoli, Italy
(3) Università' di Napoli Parthenope, via Acton 38, 80133 Napoli, Italy


Thanks to the recent advances of space technology and to the current planning for the next missions involving constellations of cooperative, multi-purpose satellites, the number of images relative to the same scene is expected to significantly grow in the next future. According to the possibility of collecting different data sets (multidimensional data) acquired with different acquisition characteristics (view angles, times, frequency, modes, etc.), the need of developing suitable processing techniques, able of “optimally” exploiting the available multiple information, is considerably growing. Multiple channel SAR imaging is a natural extension of the well known SAR interferometry technique, where two-passes are exploited to “resolve” the scene targets in the direction orthogonal to the azimuth-range plane thus allowing reconstructing the scene Digital Elevation Model (DEM). Multipass differential SAR interferometry is a first, though important, example of how multichannel data in existing data archives can be combined to improve the imaging capability of available sensors, i.e., allowing accurate monitoring of slow ground deformations. The key point behind the work is to exploit the information acquired in multipass configuration to improve the scene range resolution by assuming the knowledge of the surface height profile. This idea has been already proposed in literature [1] for the case of two passes and flat terrain topography: the resulting problem is space invariant. We extended in [2] the problem to the case of more than two passes and relaxed the flat, or better planar, terrain topography hypothesis. In particular the problem were described in the framework of linear signal estimation problem, thus allowing the application of classical linear inversion algorithm. The theory were validated with simulated data [2]. In this contribution we first describe the results of additional experiments aimed at investigating the sensitivity and the accuracy of the technique. Moreover we address the case of real ENVISAT ASAR data and we describe the processing steps and the solution we have implemented to tackle computational limitations of reconstruction process, and thus to allow applying the technique to extended real 2D scenes.

[1] C. Prati, F. Rocca, “Improving Slant-Range Resolution With Multiple SAR Surveys”, IEEE Trans. Geosci. Remote Sens., vol. 29, pp. 135-143, January 1993

[2] G. Fornaro, V. Pascazio, G. Schirinzi, “Resolution Improvement via Multipass SAR Imaging”, IGARSS'2001, Sydney (AUS), 2001.


This work has been carried out under the ESA Category 1 Project C1P.3173.


Workshop presentation

Full paper


  Higher level                 Last modified: 07.10.03