Creating and sharing knowledge for telecommunications


Cedillo, J. ; Martin, G. Martin ; Nascimento, J.

HYPERSPECTRAL IMAGE RECONSTRUCTION FROM RANDOM PROJECTIONS ON GPU, Proc IEEE International Geoscience and Remote Sensing Symp.- IGARSS, Beijin, China, Vol. -, pp. - - -, July, 2016.

Digital Object Identifier:

Hyperspectral data compression and dimensionality reduction has received considerable interest in recent years due to the high spectral resolution of these images. Contrarily to the conventional dimensionality reduction schemes, the SpeCA methodology performs dimensionality reduction basing random projections. The SpeCA methodology has applications in Hyperspectral Compressive Sensing and also in dimensionality reduction. Due to the extremely large volumes of data collected by imaging spectrometers, high performance computing architectures are need for needed in data compression of high dimensional hyperspectral data for their use in applications under real-time constraints. In this paper we propose a parallel implementation of SpeCA on Graphics Processing Units (GPUs) using the compute unified device architecture (CUDA). The proposed implementation is performed in a pixel-by-pixel fashion using coalesced accesses to memory and exploiting shared memory to store temporary data. Furthermore, the kernels have been optimized to minimize the threads divergence, therefore, achieve high GPU occupancy. The experimental results obtained for simulated and real hyperspectral data sets reveal speedups up to 21 times, which demonstrates that the GPU implementation can significantly accelerate the methods execution over big datasets while maintaining the methods accuracy.