3737

GPU implementation of the pixel purity index algorithm for hyperspectral image analysis

Sergio Sanchez, Antonio Plaza
Hyperspectral Computing Laboratory, Department of Technology of Computers and Communications, University of Extremadura, Avda. de la Universidad s/n, E-10071 Caceres, Spain
IEEE International Conference on Cluster Computing Workshops and Posters (CLUSTER WORKSHOPS), 2010

@conference{sanchez2010gpu,

   title={GPU implementation of the pixel purity index algorithm for hyperspectral image analysis},

   author={S{‘a}nchez, S. and Plaza, A.},

   booktitle={Cluster Computing Workshops and Posters (CLUSTER WORKSHOPS), 2010 IEEE International Conference on},

   pages={1–7},

   organization={IEEE}

}

Download Download (PDF)   View View   Source Source   

761

views

Hyperspectral imaging is a new technique in remote sensing that generates images with hundreds of spectral bands, at different wavelength channels, for the same area on the surface of the Earth. The price paid for such a wealth of spectral information is the enormous amounts of data to be processed. In recent years, several efforts have been directed towards the incorporation of high-performance computing models in remote sensing missions. For this purpose, graphics processing units (GPUs) have emerged as a very interesting type of hardware architecture in hyperspectral image processing due to its low weight and compact size, which allows for on-board data processing. In this paper, we develop an innovative GPU implementation of a standard hyperspectral image processing algorithm called pixel purity index (PPI) and utilized, among others, in commercial software tools such as ITTVIS Environment for Visualizing Images (ENVI) software originally developed by Analytical Imaging and Geophysics (AIG), one of the most popular tools currently available for processing remotely sensed data. The algorithm has been implemented using the compute device unified architecture (CUDA), and tested on the NVidia Tesla C1060 architecture, achieving a significant performance increase in the analysis of both synthetic and real hyperspectral data.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2017 hgpu.org

All rights belong to the respective authors

Contact us: