Classifying Complex Mountainous Forests with L-Band SAR and Landsat Data Integration: A Comparison among Different Machine Learning Methods in the Hyrcanian Forest

Remote Sens. 2014, 6(5), 3624-3647; doi:10.3390/rs6053624

Author

Sara Attarchi ^1,*  and Richard Gloaguen ^1,2

¹

Remote Sensing Group, Institute of Geology, TU Bergakademie Freiberg (TUBAF), Bernhard-von-Cotta-Str. 2, D-09599 Freiberg, Germany

²

Remote Sensing Group, Helmholtz Institute Freiberg of Resource Technology, Halsbruecker Str. 34, D-09599 Freiberg, Germany

^*

Author to whom correspondence should be addressed.

Received: 20 January 2014 / Revised: 9 April 2014 / Accepted: 15 April 2014 / Published: 25 April 2014

View Full-Text   |     Download PDF [1756 KB, uploaded 19 June 2014]   |    

 Browse Figures

 

Abstract 

Forest environment classification in mountain regions based on single-sensor remote sensing approaches is hindered by forest complexity and topographic effects. Temperate broadleaf forests in western Asia such as the Hyrcanian forest in northern Iran have already suffered from intense anthropogenic activities. In those regions, forests mainly extend in rough terrain and comprise different stand structures, which are difficult to discriminate. This paper explores the joint analysis of Landsat7/ETM+, L-band SAR and their derived parameters and the effect of terrain corrections to overcome the challenges of discriminating forest stand age classes in mountain regions. We also verified the performances of three machine learning methods which have recently shown promising results using multisource data; support vector machines (SVM), neural networks (NN), random forest (RF) and one traditional classifier (i.e., maximum likelihood classification (MLC)) as a benchmark. The non-topographically corrected ETM+ data failed to differentiate among different forest stand age classes (average classification accuracy (OA) = 65%). This confirms the need to reduce relief effects prior data classification in mountain regions. SAR backscattering alone cannot properly differentiate among different forest stand age classes (OA = 62%). However, textures and PolSAR features are very efficient for the separation of forest classes (OA = 82%). The highest classification accuracy was achieved by the joint usage of SAR and ETM+ (OA = 86%). However, this shows a slight improvement compared to the ETM+ classification (OA = 84%). The machine learning classifiers proved t o be more robust and accurate compared to MLC. SVM and RF statistically produced better classification results than NN in the exploitation of the considered multi-source data. View Full-Text

Keywords: Landsat;  ALOS/PALSAR;  L-band;  maximum likelihood classification;  support vector machines;  neural networks; random forest;  topographic effects;  Hyrcanian mountainous forest;  Iran

This is an open access article distributed under the Creative Commons Attribution License (CC BY 3.0).

For further details log on website :
http://www.mdpi.com/2072-4292/6/5/3624