GPU Accelerated FFT-Based Registration of Hyperspectral Scenes

TítuloGPU Accelerated FFT-Based Registration of Hyperspectral Scenes
AutoresÁlvaro Ordóñez, Francisco Argüello and Dora B. Heras
TipoArtículo de revista
Fonte IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, IEEE, Vol. 10, No. 11, pp. 4869-4878 , 2017.
RankRanked Q1 in Computers in Earth Sciences by SJR
ISSN1939-1404
DOI10.1109/JSTARS.2017.2734052
AbstractRegistration is a fundamental previous task in many applications of hyperspectrometry. Most of the algorithms developed are designed to work with RGB images and ignore the execution time. This paper presents a phase correlation algorithm on GPU to register two remote sensing hyperspectral images. The proposed algorithm is based on principal component analysis, multilayer fractional Fourier transform, combination of log-polar maps, and peak processing. It is fully developed in CUDA for NVIDIA GPUs. Different techniques such as the efficient use of the memory hierarchy, the use of CUDA libraries, and the maximization of the occupancy have been applied to reach the best performance on GPU. The algorithm is robust achieving speedups in GPU of up to 240.6×.
Palabras chaveGraphics processing units, Correlation, Hyperspectral imaging, Principal component analysis, Algorithm design and analysis, Fourier transforms