Title :
Parallel implementation of vertex component analysis for hyperspectral endmember extraction
Author :
Alves, José M Rodríguez ; Nascimento, José M P ; Bioucas-Dias, José M. ; Silva, Vítor ; Plaza, Antonio
Author_Institution :
Inst. de Telecomun., Lisbon, Portugal
Abstract :
Vertex component analysis (VCA) has become a very popular and useful tool to linear unmix large hyperspectral datasets without the use of any a priori knowledge of the constituent spectra. Although VCA is fast method, many hyperspectral imagery applications require a response in real time or near-real time. This paper proposes two different optimizations for accelerating the computational performance of VCA: the first one focus a parallel implementation based on graphics computing units (GPUs) to alleviate the VCA computational burden; The second one is focused on the development of a strategy to remove a large proportion of mixed pixels that play no effect on the VCA functioning. Experiments are conducted using simulated and real hyperspectral datasets. These results reveal considerable acceleration factors, which satisfies the real-time constraints given by the data acquisition rate.
Keywords :
data acquisition; geophysical image processing; geophysical techniques; graphics processing units; parallel processing; spectral analysis; GPU; VCA computational performance; acceleration factor; data acquisition rate; graphics computing units; hyperspectral endmember extraction; hyperspectral imagery application; large hyperspectral dataset; linear unmixing; mixed pixels; parallel implementation; real-time constraint; vertex component analysis; Acceleration; Algorithm design and analysis; Graphics processing units; Hyperspectral imaging; Real-time systems; Vectors; Endmember Extraction; Graphics Processing Unit; Hyperspectral Unmixing; Parallel Methods; Vertex Component Analysis;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2012 IEEE International
Conference_Location :
Munich
Print_ISBN :
978-1-4673-1160-1
Electronic_ISBN :
2153-6996
DOI :
10.1109/IGARSS.2012.6350513