Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/44974
Title: Towards the acceleration of sequential endmember extraction algorithms for time critical applications
Authors: Guerra, Raul 
Lopez, Sebastian 
Callico, Gustavo M. 
Lopez, Jose F. 
Sarmiento, Roberto 
UNESCO Clasification: 3307 Tecnología electrónica
Keywords: Hyperspectral imaging
Algorithm design and analysis
Acceleration
Classification algorithms
Issue Date: 2013
Journal: Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 
Conference: 5th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2013 
Abstract: This paper presents a new strategy in order to accelerate the execution of sequential endmember extraction algorithms without compromising their performance in terms of the accuracy of the estimated endmembers. In particular, our proposal takes advantage of the correlation between pixels located in adjacent spatial positions as well as of the information provided by the dimensionality reduction step that takes place priors to the endmember extraction itself. The results demonstrate that when the proposed strategy is applied to the well-known Vertex Component Analysis (VCA) sequential algorithm, almost half of the computing time is saved with negligible variations in the quality of the endmembers extracted. Moreover, this is achieved with independence of the amount of noise and/or the number of endmembers of the hyperspectral image under processing.
URI: http://hdl.handle.net/10553/44974
ISBN: 9781509011193
ISSN: 2158-6276
DOI: 10.1109/WHISPERS.2013.8080619
Source: Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing[ISSN 2158-6276], p. 1-4
Appears in Collections:Actas de congresos
Thumbnail
Adobe PDF (217,97 kB)
Show full item record

Page view(s)

36
checked on Mar 1, 2021

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.