Library Subscription: Guest
Begell Digital Portal Begell Digital Library eBooks Journals References & Proceedings Research Collections
Journal of Automation and Information Sciences

Impact factor: 0.024

ISSN Print: 1064-2315
ISSN Online: 2163-9337

Volumes:
Volume 49, 2017 Volume 48, 2016 Volume 47, 2015 Volume 46, 2014 Volume 45, 2013 Volume 44, 2012 Volume 43, 2011 Volume 42, 2010 Volume 41, 2009 Volume 40, 2008 Volume 39, 2007 Volume 38, 2006 Volume 37, 2005 Volume 36, 2004 Volume 35, 2003 Volume 34, 2002 Volume 33, 2001 Volume 32, 2000 Volume 31, 1999 Volume 30, 1998 Volume 29, 1997 Volume 28, 1996

Journal of Automation and Information Sciences

DOI: 10.1615/JAutomatInfScien.v44.i5.70
pages 67-80

Efficiency Assessment of Different Approaches to Crop Classification Based on Satellite and Ground Observations

Javier Gallego
European Commission Joint Research Center (JRC), Ispra (Italy)
Alexey N. Kravchenko
Institute of Space Research of National Academy of Sciences of Ukraine and State Space Agency of Ukraine, Kiev
Nataliya N. Kussul
Institute of Space Research of National Academy of Sciences of Ukraine and State Space Agency of Ukraine, Kiev, Ukraine
Sergey V. Skakun
Institute of Space Research of National Academy of Sciences of Ukraine and National Space Agency of Ukraine, Kiev, Ukraine
Andrey Yu. Shelestov
National University of Life and Environmental Sciences of Ukraine, Kiev
Yulia A. Grypych
Institute of Space Research of National Academy of Sciences of Ukraine and State Space Agency of Ukraine, Kiev

ABSTRACT

A problem of crop plants classification for three regions of Ukraine with an area of 78.500 km2 is considered. Classification is carried out using not a single satellite but a time series of satellite images. The used satellite data are characterized by different spatial resolution and temporal characteristics. By example of this problem we assessed the efficiency of different classification algorithms (neural networks, decision trees and support vector machines) for substantially different data levels (of training and testing samples) both for extremely large data sets and under the condition of their lack (absence).