| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 113.92 KB | Adobe PDF |
Advisor(s)
Abstract(s)
In this paper, a parallel implementation of a previously constrained hyperspectral coded aperture (CHYCA) algorithm for compressive sensing on graphics processing units (GPUs) is proposed. CHYCA method combines the ideas of spectral unmixing and compressive sensing exploiting the high spatial correlation that can be observed in the data and the generally low number of endmembers needed in order to explain the data. The performance of CHYCA relies which does not depend on the tuning of a regularization parameter, which is a time consuming task offering good performance compared with a previously hyperspectral coded aperture (HYCA) method. The proposed implementation exploits the GPU architecture at low level, thus taking full advantage of the computational power of GPUs using shared memory and coalesced accesses to memory. Experimental results using simulated data reveals speedups up to 56 times, with regards to serial implementation.
Description
Keywords
Hyperspectral imaging Compressive sensing (CS) Coded aperture Graphics processing units (GPUS)
Citation
BARNABÉ, Sérgio; [et al] – GPU implementation of a constrained hyperspectral coded aperture algorithm for compressive sensing. In 2015 7th Workshop on Hyperspectral Image and Signal Processing - Evolution in Remote Sensing (WHISPERS). Tokyo, Japan: IEEE, 2015. ISBN 978-1-4673-9015-6. Pp. 1-4
Publisher
Institute of Electrical and Electronics Engineers