SAR Image Reconstruction Method of Incomplete Raw Data Based on Matrix Completion

Rahmat Arief (2016) SAR Image Reconstruction Method of Incomplete Raw Data Based on Matrix Completion. Proceedings The 2nd International Conference of Indonesian Society for Remote Sensing 2016, 47. pp. 78-88.

Prosiding_Rahmat Arief_Pustekdata_2016.pdf

Download (2MB) | Preview


This paper discusses a new approach of SAR image reconstruction based on matrix completion from afew number of samples of the received SAR signal required by the Nyquist theorem in both the azimuthand range. The reconstruction method has goal to fill the missing information in the raw data caused byunder sampling based on matrix completion of Fourier coefficients. Experiments were performed on the simulated point target and the real RADARSAT1data in noise freeand noisy condition. Compared with the conventional method as range doppler algorithm, the proposed method presented slightly better results in suppressing of side lobe and canmaintain resolution SAR imagery, but with the advantage with a few number of measurements. Asexpected, the larger the number of measurements, a target SAR image can be reconstructed rarelybetter. The experiments on the RADARSAT1data showed the almost same performance as in thesimulation on the target points.

Item Type: Article
Additional Information: ISBN 978-602-73620-1-7
Uncontrolled Keywords: SAR, compressive sensing, matrix completion, filling missing, Radarsat-1
Subjects: Teknologi Penginderaan Jauh > Pengelolaan dan Pengembangan > Citra Satelit
Divisions: Deputi Penginderaan Jauh > Pusat Teknologi dan Data Penginderaan Jauh
Depositing User: Dinar Indrasasi
Date Deposited: 18 Apr 2021 06:08
Last Modified: 13 Oct 2021 10:09

Actions (login required)

View Item View Item