Efficient parallel implementation of polarimetric synthetic aperture radar data processing

Sergio S. Martinez; Prashanth R. Marpu; Antonio J. Plaza

doi:10.1117/12.2067414

Translator Disclaimer

15 October 2014 Efficient parallel implementation of polarimetric synthetic aperture radar data processing

Sergio S. Martinez, Prashanth R. Marpu, Antonio J. Plaza

Author Affiliations +

Proceedings Volume 9247, High-Performance Computing in Remote Sensing IV; 92470K (2014) https://doi.org/10.1117/12.2067414
Event: SPIE Remote Sensing, 2014, Amsterdam, Netherlands

Abstract

This work investigates the parallel implementation of polarimetric synthetic aperture radar (POLSAR) data processing chain. Such processing can be computationally expensive when large data sets are processed. However, the processing steps can be largely implemented in a high performance computing (HPC) environ- ment. In this work, we studied different aspects of the computations involved in processing the POLSAR data and developed an efficient parallel scheme to achieve near-real time performance. The algorithm is implemented using message parsing interface (MPI) framework in this work, but it can be easily adapted for other parallel architectures such as general purpose graphics processing units (GPGPUs).

Citation Download Citation

Sergio S. Martinez, Prashanth R. Marpu, and Antonio J. Plaza "Efficient parallel implementation of polarimetric synthetic aperture radar data processing", Proc. SPIE 9247, High-Performance Computing in Remote Sensing IV, 92470K (15 October 2014); https://doi.org/10.1117/12.2067414

ACCESS THE FULL ARTICLE