Parallel Hyperspectral Unmixing on GPUs
; Alves, J.M.R.A
; Plaza, A.
IEEE Geoscience and Remote Sensing Letters Vol. 11, Nº 3, pp. 666 - 670, March, 2014.
ISSN (print): 1545-598X
Scimago Journal Ranking: 1,41 (in 2014)
Digital Object Identifier: 10.1109/LGRS.2013.2274328
This work presents a new parallel method for hyperspectral unmixing composed by the efficient combination of two popular methods: vertex component analysis (VCA) and sparse unmixing by variable splitting and augmented Lagrangian (SUNSAL). First, VCA extracts the endmembers signatures and
then SUNSAL is used to estimate the abundance fractions. Both
techniques are highly parallelizable, which reduces significantly
the computing time. A design for commodity graphics processing
units (GPUs) of the two methods is presented and evaluated. Experimental results obtained for simulated and real hyperspectral datasets reveal speedups up to 100 times, which grants realtime response required by many remotely sensed hyperspectral applications.