Please use this identifier to cite or link to this item:
Title: Fuel type mapping in the mediterranean region of north Lebanon using object-based image analysis of ASTER imagery
Authors: Mitri, George 
Nader, Manal 
Salloum, Liliane
Affiliations: Institute of Environment 
Institute of Environment 
Keywords: Fuel type mapping
Object-Based Image Analysis
Issue Date: 2011
Publisher: Office of the European Union
Part of: Advances in Remote Sensing and GIS applications in Forest Fire Management From local to global assessments
Start page: 39
End page: 43
Conference: International EARSeL FF-SIG Workshop (8th : 21-22 October 2011 : Stresa, Italy) 
Forests and Other Wooded Land in Lebanon are a unique feature in the semi-arid environment of the Eastern Mediterranean. Until 2006, they covered approximately 24% of the overall area of Lebanon. The forests are di-vided into three main classes, namely mixed forest, broadleaves, and coniferous. The Other Wooded Land is di-vided into the following classes: coniferous shrubs, broadleaved shrub, mixed shrublands and grassland with trees. Like other Euro-Mediterranean countries, forest fires have been especially damaging in Lebanon in recent years, representing one of the most important elements that contributes to the destruction of Lebanons natural resources. Most recently, a National Strategy for forest fire management was officially endorsed by the Govern-ment of Lebanon. One of the main activities of the National Strategy is to develop a fuel management plan aim-ing at reducing the highly flammable biomass. Most commonly, fuel maps in the Mediterranean are generated from remote sensing data, mainly, medium resolution sensors such as Landsat data and Very High Resolution sensors such as IKONOS data. The aim of this work was to present a classification approach to generate fuel type maps in the Eastern Mediterranean using ASTER imagery. The Prometheus fuel type classification system which is adapted to the ecological characteristics of the European Mediterranean basin was adopted. Field visits for the recognition of different fuel types were conducted. The field data were used as ground-truth dataset to train the classification model and to assess the classification results obtained for the study area. The Object-Based Image Analysis (OBIA) approach was used for fuel type mapping. This involved three steps, namely, image segmentation, object training and object classification. The process resulted in the separation of six fuel type classes. Varying degree of accuracy levels among the different fuel type classes was preliminary achieved. The results showed that the u.
Open URL: Link to full text
Type: Conference Paper
Appears in Collections:Institute of Environment

Show full item record

Record view(s)

checked on Dec 2, 2021

Google ScholarTM


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