Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/44973
Title: Random forest training stage acceleration using graphics processing units
Authors: Hernández Guedes, Abián 
Fabelo, Himar 
Ortega, Samuel 
Báez Quevedo, Abelardo 
Callicó, Gustavo M. 
Sarmiento Rodríguez, Roberto 
UNESCO Clasification: 3308 Ingeniería y tecnología del medio ambiente
Keywords: Hyperspectral Imaging
Supervised Learning
High Performance Computing
Graphical Processing Units
Random Forest
Issue Date: 2018
Journal: Proceedings (Conference on Design of Circuits and Integrated Systems) 
Conference: 32nd Conference on Design of Circuits and Integrated Systems, DCIS 2017 
Abstract: Graphics Processing Units (GPUs) are platforms very appropriated to accelerate processes with high computational load, like the supervised classification of hyperspectral images. The supervised classifier Random Forest has proved to be a good candidate to classify hyperspectral images and currently constitutes an emerging technology for medical diagnosis. The objective of this paper is focused in the Random Forest training phase acceleration using GPUs, starting from an efficient CPU implementation. For some applications, it is necessary to refine the classification model depending on the new acquired samples. In this paper are presented solutions for two bottlenecks identified in the training stage in order to accelerate the algorithm. The different solutions for the bottlenecks provided in this research study have demonstrated that GPU implementation is a promising technique to generate models in shorter time. With this implementation it is possible to achieve the training process in real-time.
URI: http://hdl.handle.net/10553/44973
ISBN: 978-1-5386-5108-7
ISSN: 2471-6170
DOI: 10.1109/DCIS.2017.8311636
Source: 2017 32nd Conference on Design of Circuits and Integrated Systems, DCIS 2017 - Proceedings,v. 2017-November, p. 1-6
Appears in Collections:Actas de congresos
Show full item record

SCOPUSTM   
Citations

2
checked on Mar 24, 2024

WEB OF SCIENCETM
Citations

1
checked on Feb 25, 2024

Page view(s)

93
checked on Jul 22, 2023

Google ScholarTM

Check

Altmetric


Share



Export metadata



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